<?xml version='1.0' encoding='utf-8'?>
<!DOCTYPE rfc [
  <!ENTITY nbsp    "&#160;">
  <!ENTITY zwsp   "&#8203;">
  <!ENTITY nbhy   "&#8209;">
  <!ENTITY wj     "&#8288;">
]>
<?xml-stylesheet type="text/xsl" href="rfc2629.xslt" ?>
<!-- generated by https://github.com/cabo/kramdown-rfc version 1.7.43 (Ruby 3.2.3) -->
<rfc xmlns:xi="http://www.w3.org/2001/XInclude" ipr="trust200902" docName="draft-ietf-cats-metric-definition-15" category="std" submissionType="IETF" tocInclude="true" sortRefs="true" symRefs="true" version="3">
  <!-- xml2rfc v2v3 conversion 3.35.0 -->
  <front>
    <title abbrev="CATS Metrics">CATS Metrics Definition</title>
    <seriesInfo name="Internet-Draft" value="draft-ietf-cats-metric-definition-15"/>
    <author initials="Y." surname="Kehan" fullname="Kehan Yao">
      <organization>China Mobile</organization>
      <address>
        <postal>
          <country>China</country>
        </postal>
        <email>yaokehan@chinamobile.com</email>
      </address>
    </author>
    <author initials="C." surname="Li" fullname="Cheng Li">
      <organization>Huawei Technologies</organization>
      <address>
        <postal>
          <country>China</country>
        </postal>
        <email>c.l@huawei.com</email>
      </address>
    </author>
    <author initials="L. M." surname="Contreras" fullname="L. M. Contreras">
      <organization>Telefonica</organization>
      <address>
        <email>luismiguel.contrerasmurillo@telefonica.com</email>
      </address>
    </author>
    <author initials="J." surname="Ros-Giralt" fullname="Jordi Ros-Giralt">
      <organization>Qualcomm Europe, Inc.</organization>
      <address>
        <email>jros@qti.qualcomm.com</email>
      </address>
    </author>
    <author initials="G." surname="Zeng" fullname="Guanming Zeng">
      <organization>Huawei Technologies</organization>
      <address>
        <postal>
          <country>China</country>
        </postal>
        <email>zengguanming@huawei.com</email>
      </address>
    </author>
    <date year="2026" month="November" day="16"/>
    <area>Routing</area>
    <workgroup>Computing-Aware Traffic Steering</workgroup>
    <keyword>CATS, metrics</keyword>
    <abstract>
      <?line 110?>

<t>Computing-Aware Traffic Steering (CATS) is a traffic engineering approach that optimizes the steering of traffic to a service instance by considering the dynamic state of computing and network resources. To
enable such decisions, CATS components exchange metrics that describe resource conditions affecting service instance selection. This document focuses on compute and communication metrics for CATS and defines a
hierarchical abstraction of these metrics to improve interoperability, scalability, and operational simplicity. It does not aim to standardize raw infrastructure (Level 0) metrics; instead, it specifies higher-level representations that can be derived from raw measurements using aggregation and normalization functions.</t>
    </abstract>
  </front>
  <middle>
    <?line 116?>

<section anchor="introduction">
      <name>Introduction</name>
      <t>Service providers are deploying computing capabilities across the network for hosting applications such as distributed AI workloads, AR/VR and driverless vehicles, among others. In these deployments, multiple service instances are replicated across various sites to ensure sufficient capacity for maintaining the required Quality of Experience (QoE) expected by the application. To support the selection of these instances, a framework called Computing-Aware Traffic Steering (CATS) is introduced in <xref target="RFC10053"/>.</t>
      <t>CATS is a traffic engineering approach that optimizes the steering of traffic to a given service instance by considering the dynamic nature of computing and network resources. To achieve this, CATS components require performance metrics for both communication and compute resources. Since these resources are deployed by multiple providers, standardized metrics are essential to ensure interoperability and enable precise traffic steering decisions, thereby optimizing resource utilization and enhancing overall system performance.</t>
      <t>There are already well-defined network metrics for traffic steering, such as Traffic Engineering (TE) metrics and IGP metrics (e.g., link delay, link delay variation)<xref target="RFC7471"/>, which have been in use in network systems for a long time. In the context of CATS, computing metrics need to be introduced to enable joint TE decisions. <xref target="DMTF"/> defines some fine-grained computing metrics, such as CPU utilization, but directly using these fine-grained computing metrics lacks scalability.</t>
      <t>This document does not attempt to standardize low-level fine-grained performance metrics. Instead, it organizes computing and communication metrics into three abstraction levels and defines a metric framework based on aggregation and normalization functions. The framework specifies four categories of Level 1 metrics and a normalized Level 2 metric, balancing metric expressiveness with scalability and ease of use.</t>
    </section>
    <section anchor="conventions-definitions">
      <name>Conventions and Definitions</name>
      <t>This document uses the following terms defined in <xref target="RFC10053"/>:</t>
      <ul spacing="normal">
        <li>
          <t>Computing-Aware Traffic Steering (CATS)</t>
        </li>
        <li>
          <t>Service</t>
        </li>
        <li>
          <t>Service site</t>
        </li>
        <li>
          <t>Service contact instance</t>
        </li>
        <li>
          <t>CATS Service Contact Instance ID (CSCI-ID)</t>
        </li>
        <li>
          <t>CATS Service Metric Agent (C-SMA)</t>
        </li>
        <li>
          <t>CATS Network Metric Agent (C-NMA)</t>
        </li>
        <li>
          <t>CATS Path Selector (C-PS)</t>
        </li>
      </ul>
      <t>The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 <xref target="RFC2119"/> <xref target="RFC8174"/> when, and only when, they appear in all capitals, as shown here.</t>
    </section>
    <section anchor="design-principles">
      <name>Design Principles</name>
      <section anchor="three-level-metrics">
        <name>Three-Level Metrics</name>
        <t>As outlined in <xref target="RFC10054"/>, the resource model that defines CATS metrics MUST be scalable, ensuring that its implementation remains within a reasonable and sustainable cost. To that end, a CATS system should select the most appropriate metrics for instance selection, recognizing that different metrics may influence outcomes in distinct ways depending on the specific use case.</t>
        <t>Defining metrics requires carefully balancing multiple considerations, including metric diversity, granularity, and rate of change (e.g., update frequency or advertisement churn). An excessive number of
metrics, overly fine granularity, or high update frequency can lead to significant signaling overhead, reducing scalability of the metric distribution protocol. In contrast, metrics that are too few, too
coarse-grained, or updated too infrequently may fail to provide sufficient information to support effective operational decisions.</t>
        <t>Conceptually, it is necessary to define at least two fundamental levels of metrics: one comprising all raw metrics, and the other representing a simplified form - consisting of a single value that encapsulates the overall capability of a service instance.</t>
        <t>However, such a definition may reduce implementation flexibility across diverse CATS use cases. Implementers typically seek balanced approaches that carefully manage trade-offs among encoding complexity, accuracy, scalability, and extensibility.</t>
        <t>To ensure scalability while providing sufficient detail for effective decision-making, this document provides a definition of metrics that incorporates three levels of abstraction:</t>
        <ul spacing="normal">
          <li>
            <t><strong>Level 0: Raw metrics.</strong> These metrics are presented without abstraction, with each metric using its own unit and format as defined by the underlying resource.</t>
          </li>
          <li>
            <t><strong>Level 1: Metrics combined into categories.</strong> These metrics are derived from Level 0 metrics by applying aggregation functions and, optionally, normalization functions to form category-specific metrics, such as computing and communication.</t>
          </li>
          <li>
            <t><strong>Level 2: A single normalized metric.</strong> This metric is computed by aggregating lower-level metrics (Level 0
or Level 1) and applying normalization to produce a single, unitless Level 2 score within a defined range.</t>
          </li>
        </ul>
      </section>
      <section anchor="level-0-metrics">
        <name>Level 0: Raw Metrics</name>
        <t>Level 0 metrics represent detailed, raw measurements collected from
underlying resources. These metrics are typically service-specific and
are not abstracted.</t>
        <t>Examples of Level 0 metrics include, but are not limited to:</t>
        <ul spacing="normal">
          <li>
            <t><strong>CPU:</strong> Base frequency, boosted frequency, number of cores, core
utilization, memory bandwidth, memory capacity, memory utilization,
and power consumption.</t>
          </li>
          <li>
            <t><strong>GPU:</strong> Frequency, number of processing units, memory bandwidth,
memory capacity, memory utilization, core utilization, and power
consumption.</t>
          </li>
          <li>
            <t><strong>NPU:</strong> Computational capacity, utilization, and power consumption.</t>
          </li>
          <li>
            <t><strong>Communication:</strong> Throughput, bandwidth, link utilization, packet
loss, delay, jitter, traffic counters (bytes and packets), and other
network performance indicators. Network performance metrics and their
measurement methods are classified in <xref target="RFC7799"/>.</t>
          </li>
          <li>
            <t><strong>Storage:</strong> Available capacity, read throughput, and write throughput.</t>
          </li>
          <li>
            <t><strong>Service-specific metrics:</strong> Request rate (e.g., requests per second),
output rate (e.g., tokens per second), and other application-level
performance indicators.</t>
          </li>
        </ul>
        <t>Level 0 metrics serve as the foundational inputs for the metric
hierarchy. Some metrics are derived from monitoring systems (e.g.,
telemetry or counters), others reflect dynamic runtime state, and
others may correspond to relatively static properties of the underlying
infrastructure. These metrics provide the basic information required to
derive higher-level metrics, as described in the following sections.</t>
        <t>Level 0 metrics can be encoded and exposed using an Application Programming Interface (API), such as a RESTful API, and can be technology- and implementation-specific. Different resources can have their own metrics, each conveying unique information about their status. These metrics can generally have units, such as bits per second (bps) or floating point instructions per second (flops), or be unitless, such as CPU utilization.</t>
        <t>As examples, <xref target="RFC8911"/> and <xref target="RFC8912"/> define various network performance
metrics and their associated registries, while <xref target="DMTF"/> defines a
set of computing metrics. These Level 0 metrics are not standardized in
this document; rather, they serve as foundational inputs that can be used
within CATS to derive higher-level metrics.</t>
      </section>
      <section anchor="level-1-metrics">
        <name>Level 1: Metrics Combined in Categories</name>
        <t>Level 1 metrics are grouped into four categories: computing, communication, service, and composed, with the possibility of additional categories being defined in future specifications. For each category, a single Level 1 metric is derived through an aggregation function and, when appropriate, further normalized to
yield a unitless score reflecting the performance of the underlying resources. The Level 1 categories are described as follows:</t>
        <ul spacing="normal">
          <li>
            <t><strong>Computing:</strong> A value derived from aggregating one or more computing-related Level 0 metrics, such as CPU, GPU, and NPU utilization.</t>
          </li>
          <li>
            <t><strong>Communication:</strong> A value derived from aggregating one or more communication-related Level 0 metrics, such as communication throughput.</t>
          </li>
          <li>
            <t><strong>Service:</strong> A value derived from aggregating one or more service-related Level 0 metrics, such as tokens per second and service availability</t>
          </li>
          <li>
            <t><strong>Composed:</strong> A value derived from aggregating a combination of computing, communication, and service metrics.</t>
          </li>
        </ul>
        <t>Refer to <xref target="aggregation-function"/> and <xref target="normalization-function"/> for the definitions and examples of aggregation functions and normalization functions, respectively. Refer to <xref target="ops-considerations"/> for the default policies and guidance provided to implementations.</t>
        <t>Level 1 metrics allow to focus solely on the metric categories and their simple values, thereby avoiding the need to process solution-specific Level 0 metrics.</t>
      </section>
      <section anchor="level-2-metric">
        <name>Level 2: A Single Normalized Metric</name>
        <t>The Level 2 metric is a single, normalized score derived from lower-level metrics (Level 0 and/or Level 1) through the application of aggregation and normalization functions. Different implementations
may apply different functions to characterize the overall performance of the underlying computing and communication resources. By consolidating multiple lower-level metrics into a single score, the Level 2 metric significantly reduces the complexity associated with metric collection and distribution. <xref target="ops-considerations"/> further describes default policies for implementations.</t>
        <t>Figure 1 provides a summary of the logical relationships between metrics across the three levels of abstraction.</t>
        <figure anchor="fig-metric-levels">
          <name>Logic of CATS Metrics in levels</name>
          <artwork><![CDATA[
                                   +--------+
              Level 2 Metric:      |   M2   |
                                   +---^----+
                                       |
                         +-------------+-----------+------------+
                         |             |           |            |
                     +---+----+        |       +---+----+   +---+----+
 Level 1 Metrics:    |  M1-1  |        |       |  M1-2  |   |  M1-3  | (...)
                     +---^----+        |       +---^----+   +----^---+
                         |             |           |             |
                    +----+---+         |       +---+----+        |
                    |        |         |       |        |        |
                 +--+---+ +--+---+ +---+--+ +--+---+ +--+---+ +--+---+
 Level 0 Metrics:| M0-1 | | M0-2 | | M0-3 | | M0-4 | | M0-5 | | M0-6 | (...)
                 +------+ +------+ +------+ +------+ +------+ +------+

]]></artwork>
        </figure>
      </section>
    </section>
    <section anchor="cats-metric-framework">
      <name>CATS Metrics Framework and Specification</name>
      <t>The CATS metrics framework defines how metrics are encoded and transmitted over the network. The representation should be flexible enough to accommodate various types of metrics along with their respective units and precision levels, yet simple enough to enable easy implementation and deployment across heterogeneous edge environments.</t>
      <t>The design of the CATS metrics framework is guided by the following
principles:</t>
      <ul spacing="normal">
        <li>
          <t><strong>Semantic granularity and extensibility:</strong> The framework adopts a
layered abstraction of metrics to balance expressiveness and
scalability. By organizing metrics into multiple levels of increasing
abstraction (e.g., raw, aggregated, and normalized), it enables
implementations to select the appropriate level of detail for their
use case. This approach allows fine-grained metrics to be preserved at
lower levels while exposing more compact and semantically meaningful
representations at higher levels. In addition, the layered design
supports extensibility by allowing new metrics and categories to be
introduced without disrupting existing deployments.</t>
        </li>
        <li>
          <t><strong>Interoperability and flexibility:</strong> The framework allows
implementation-specific aggregation and normalization functions to
accommodate diverse deployment scenarios and operational objectives within a single administrative domain.
At the same time, it defines common metric structures and introduces
default policies to guide interpretation, ensuring a consistent
understanding of metrics across vendors. This combination
of flexibility and guidance enables interoperability while preserving
innovation and adaptability in metric computation and usage.</t>
        </li>
        <li>
          <t><strong>Metric provenance and transparency:</strong> The framework explicitly captures the
origin and context of metrics by introducing a "Source" field, following the
model defined in <xref target="RFC9439"/>. This field distinguishes whether a metric
value is derived from direct measurement, estimation, aggregation, or
normalization. By identifying the source of each metric, the framework
improves transparency and enables implementations to better assess the
reliability, accuracy, and semantics of the reported values.</t>
        </li>
      </ul>
      <section anchor="cats-metric-fields">
        <name>CATS Metric Fields</name>
        <t>Each CATS metric is expressed as a structured set of fields, with each field describing a specific property of the metric. The following definition introduces the fields used in the CATS metric representations.</t>
        <ul spacing="normal">
          <li>
            <t><strong>Metric_Type</strong>: This field specifies the category or kind of CATS metric being reported, such as computational resources, storage capacity, or network bandwidth. It acts as a label that enables network devices to identify the purpose of the metric.</t>
          </li>
          <li>
            <t><strong>Level</strong>: This field specifies the level at which the metric is measured. It is used to categorize the metric based on its granularity and scope. There are only three valid metric levels defined in  <xref target="three-level-metrics"/>. This field can take three values: 0 for Level 0, 1 for Level 1, and 2 for Level 2.</t>
          </li>
          <li>
            <t><strong>Measurement_Method</strong>: This optional field indicates the measurement method of the metric as classified in <xref target="RFC7799"/> ('active', 'passive', or 'hybrid'). Registry entries in <xref target="cats-metrics-registry"/> are separated by measurement method. When this field is present, its value must match the measurement method of the selected registry entry. For Level 1 and Level 2 metrics, the measurement method is inherited from the underlying OAM and telemetry methods used to obtain the contributing inputs. Note that if an active measurement protocol (such as TWAMP <xref target="RFC5357"/> or STAMP <xref target="RFC8762"/>) is used solely as a transport mechanism to carry telemetry data collected by other means without performing direct active measurement, it does not alter the measurement method classification of the transported metrics.</t>
          </li>
          <li>
            <t><strong>Format</strong>: This field indicates the data encoding format of the metric, such as uint, ieee_754_float.</t>
          </li>
          <li>
            <t><strong>Length</strong>: This field indicates the size of the value field measured in octets (bytes). It specifies how many bytes are used to store the value of the metric. The length field is important for memory allocation and data handling, ensuring that the value is stored and retrieved correctly.</t>
          </li>
          <li>
            <t><strong>Unit</strong>: This field defines the measurement units for the metric, such as hertz (Hz) for frequency, bytes (B) for data size, or bits per seconds (bps) for data transfer rate. It is usually associated with the metric to provide context for the value.</t>
          </li>
          <li>
            <t><strong>Source</strong>: This field describes the origin of the information used to obtain the metric. It may include one or more of the following non-mutually exclusive values:  </t>
            <ul spacing="normal">
              <li>
                <t>'nominal'. Similar to <xref target="RFC9439"/>, "a 'nominal' metric indicates that the metric value is statically configured by the underlying devices.  For example, bandwidth can indicate the maximum transmission rate of the involved device.</t>
              </li>
              <li>
                <t>'estimation'. The 'estimation' source indicates that the metric value is computed through an estimation process.</t>
              </li>
              <li>
                <t>'directly measured'. This source indicates that the metric is obtained directly from the underlying device and it is not estimated.</t>
              </li>
              <li>
                <t>'normalization'. The 'normalization' source indicates that the metric value is normalized. This type of metrics does not have units. This document specifies that the normalized value range for each metric is 0 to 10, where 0 indicates the poorest compute/composed capability, and 10 indicates the optimal compute/composed capability.</t>
              </li>
              <li>
                <t>'aggregation'. This source indicates that the metric value is obtained by using an aggregation function.</t>
              </li>
            </ul>
            <t>
Nominal metrics have inherent physical meanings and specific units without any additional processing. Aggregated metrics may or may not have physical meanings, but they retain their significance relative to the directly measured metrics. Normalized metrics, on the other hand, might have physical meanings but lack units.</t>
          </li>
          <li>
            <t><strong>Statistics</strong>: This field provides additional details about the metrics, particularly if there is any pre-computation performed on the metrics before they are collected. This field is optional. It is useful for services that require specific statistics for service instance selection. The 'Statistics' field must be used together with the 'Measurement_Window' parameter to indicate the sampling time interval. There are four kinds of statistics:  </t>
            <ul spacing="normal">
              <li>
                <t>'max'. The maximum value of the data collected over the intervals.</t>
              </li>
              <li>
                <t>'min'. The minimum value of the data collected over the intervals.</t>
              </li>
              <li>
                <t>'mean'. The average value of the data collected over the intervals.</t>
              </li>
              <li>
                <t>'cur'. The current value of the data collected.</t>
              </li>
            </ul>
          </li>
          <li>
            <t><strong>Value</strong>: This field represents the actual numerical value of the metric being measured. It provides the specific data point for the metric in question.</t>
          </li>
          <li>
            <t><strong>Observation_Time</strong>: This field indicates the instant to which the value refers, expressed as a date and time in the format defined in <xref target="RFC3339"/>. This field is optional.</t>
          </li>
          <li>
            <t><strong>Validity_Interval</strong>: This field indicates the period, beginning at 'Observation_Time', during which the metric value remains usable for instance selection. This field is optional, and where it is absent, the usable lifetime of the value is determined by local policy.</t>
          </li>
        </ul>
        <t>The value assignment and encoding rules for these fields are specified in Section <xref target="level-metric-representations"/>.</t>
      </section>
      <section anchor="aggregation-normalization-functions">
        <name>Aggregation and Normalization Functions</name>
        <t>In the context of CATS metric processing, aggregation and normalization are two fundamental operations that transform raw and derived metrics into forms suitable for decision-making and comparison across heterogeneous systems.</t>
        <section anchor="aggregation-function">
          <name>Aggregation</name>
          <t>Aggregation functions combine multiple values into a single representative value. Aggregation functions can be applied at all metric levels. This document supports the spatial aggregation and temporal aggregation that are defined in <xref target="RFC5835"/> and <xref target="RFC5644"/>, and further defines cross-category aggregation which can aggregate metrics from different types into a single value. The following are aggregation examples supported by CATS:</t>
          <ul spacing="normal">
            <li>
              <t>Spatial or temporal aggregation of multiple metrics of the same type to produce a derived metric. In this case, because the input metrics are homogeneous, the resulting metric may retain the same units as the inputs. For example, CPU utilization measurements (expressed in percentage) collected from multiple service instances (spatial aggregation) or averaged over consecutive time intervals (temporal aggregation) can be aggregated to produce a representative CPU utilization metric. Such aggregation concepts are consistent with those described in <xref target="RFC5835"/>.</t>
            </li>
            <li>
              <t>Aggregation of multiple metrics of different types to produce a higher-level metric that captures combined behavior across resource dimensions. In this case, because the input metrics use different units, the resulting metric cannot retain physical units and must be expressed as a unitless value. For example, CPU capacity (expressed in Hz) and available memory (expressed in bytes) can be combined through aggregation to generate a single computing-time metric that characterizes overall processing capability.</t>
            </li>
          </ul>
          <t>Some common aggregation functions include:</t>
          <ul spacing="normal">
            <li>
              <t>Mean: Computes the arithmetic mean of a set of input values.</t>
            </li>
            <li>
              <t>Minimum / Maximum: Selects the lowest or highest value from a set of input values.</t>
            </li>
            <li>
              <t>Weighted average: Computes an average by applying weights to individual values according to their relative importance or priority.</t>
            </li>
          </ul>
          <t>Aggregation functions are not standardized in this document. They are implementation-specific and controlled by operator policies.</t>
          <figure anchor="fig-agg-funct">
            <name>Aggregation function</name>
            <artwork><![CDATA[
    +-----------+     +-------------------+
    | Metric 1  |---->|                   |
    +-----------+     |    Aggregation    |     +------------+
           ...        |     Function      |---->| Metric n+1 |
    +-----------+     |                   |     +------------+
    | Metric n  |---->|                   |
    +-----------+     +-------------------+

    Input: Multiple values              Output: A single value

]]></artwork>
          </figure>
        </section>
        <section anchor="normalization-function">
          <name>Normalization</name>
          <t>Normalization functions convert a metric value (with or without units) into a unitless normalized score. Normalized metrics facilitate composite scoring and ranking, and can be used to produce Level 1 and Level 2 metrics. The following are normalization examples supported by CATS:</t>
          <ul spacing="normal">
            <li>
              <t>Normalizing a single Level 0 metric to generate a Level 1 or Level 2 normalized metric;</t>
            </li>
            <li>
              <t>Normalizing the output of aggregating multiple Level 0 metrics, to generate a Level 1 normalized metric.</t>
            </li>
          </ul>
          <t>Normalization functions are commonly used to transform metric values into a bounded range (e.g., an integer scale from 0 to 10) using techniques such as sigmoid function and min-max scaling <xref target="Min-max-sigmoid"/>:</t>
          <ul spacing="normal">
            <li>
              <t>Sigmoid function: Smoothly maps input values to a bounded range.</t>
            </li>
            <li>
              <t>Min-max scaling: Rescales values based on known minimum and maximum bounds.</t>
            </li>
          </ul>
          <t>These normalization functions are also not standardized in this document. They are implementation-specific and controlled by operator policies.</t>
          <figure anchor="fig-norm-funct">
            <name>Normalization function</name>
            <artwork><![CDATA[
  +----------+     +------------------------+     +----------+
  | Metric 1 |---->| Normalization Function |---->| Metric 2 |
  +----------+     +------------------------+     +----------+

  Input:  Value with or without units         Output: Unitless value
]]></artwork>
          </figure>
        </section>
      </section>
      <section anchor="level-metric-representations">
        <name>Level Metric Representations</name>
        <t>This section specifies the representation format and constraints for
Level 1 and Level 2 metrics, ensuring consistent encoding and
interoperability across implementations.</t>
        <section anchor="level-0-representations">
          <name>Level 0 Metrics</name>
          <t>Level 0 metrics are raw metrics that are not standardized in this
document. See <xref target="appendix-level-0"/> for examples of Level 0 metrics
defined in the compute and communication industries and by other
standardization organizations such as the <xref target="DMTF"/>.</t>
        </section>
        <section anchor="level-1-representations">
          <name>Level 1 Metrics</name>
          <t>Level 1 metrics are derived from Level 0 metrics through the application
of aggregation functions and, when appropriate, normalization functions.
Depending on how they are formed, Level 1 metrics MAY retain physical
units inherited from their inputs or MAY be expressed as unitless values.</t>
          <t>Level 1 metrics are organized into semantic categories such as computing,
communication, service, and composed metrics. This categorization
provides context and meaning to the resulting metrics and enables
consistent interpretation across implementations.</t>
          <t>The 'Source' field indicates how the metric value is derived. For Level 1
metrics, typical values include:</t>
          <ul spacing="normal">
            <li>
              <t>'aggregation': The value is obtained by combining Level 0 metrics
without normalization and MAY retain a physical unit.</t>
            </li>
            <li>
              <t>'normalization': The value is mapped into a unitless score.</t>
            </li>
          </ul>
          <section anchor="level-1-computing-metrics">
            <name>Level 1 Computing Metrics</name>
            <t>The Metric Type for Level 1 computing metrics is 'level1_computing'.</t>
            <t><strong>Example A: Aggregation-derived (with units)</strong></t>
            <artwork><![CDATA[
Fields:
      Metric_type: level1_computing
      Level: Level 1
      Measurement_Method: passive
      Format: unsigned integer
      Length: two octets
      Unit: mhz
      Source: aggregation
      Value: 2400
]]></artwork>
            <t><strong>Example B: Normalized (unitless)</strong></t>
            <figure anchor="fig-level1-compute-metric">
              <name>Examples of Level 1 computing metrics</name>
              <artwork><![CDATA[
Fields:
      Metric_type: level1_computing
      Level: Level 1
      Measurement_Method: passive
      Format: unsigned integer
      Length: one octet
      Source: normalization
      Value: 5
]]></artwork>
            </figure>
          </section>
          <section anchor="level-1-communication-metrics">
            <name>Level 1 Communication Metrics</name>
            <t>The Metric Type for Level 1 communication metrics is 'level1_communication'.</t>
            <t><strong>Example A: Aggregation-derived (with units)</strong></t>
            <artwork><![CDATA[
Fields:
      Metric_type: level1_communication
      Level: Level 1
      Measurement_Method: active
      Format: unsigned integer
      Length: two octets
      Unit: mbps
      Source: aggregation
      Value: 800
]]></artwork>
            <t><strong>Example B: Normalized (unitless)</strong></t>
            <figure anchor="fig-level1-communication-metric">
              <name>Examples of Level 1 communication metrics</name>
              <artwork><![CDATA[
Fields:
      Metric_type: level1_communication
      Level: Level 1
      Measurement_Method: hybrid
      Format: unsigned integer
      Length: one octet
      Source: normalization
      Value: 1
]]></artwork>
            </figure>
          </section>
          <section anchor="level-1-service-metrics">
            <name>Level 1 Service Metrics</name>
            <t>The Metric Type for Level 1 service metrics is 'level1_service'.</t>
            <t><strong>Example A: Aggregation-derived (with units)</strong></t>
            <artwork><![CDATA[
Fields:
      Metric_type: level1_service
      Level: Level 1
      Measurement_Method: passive
      Format: unsigned integer
      Length: two octets
      Unit: rps
      Source: aggregation
      Value: 45
]]></artwork>
            <t><strong>Example B: Normalized (unitless)</strong></t>
            <figure anchor="fig-level1-service-metric">
              <name>Examples of Level 1 service metrics</name>
              <artwork><![CDATA[
Fields:
      Metric_type: level1_service
      Level: Level 1
      Measurement_Method: passive
      Format: unsigned integer
      Length: one octet
      Source: normalization
      Value: 7
]]></artwork>
            </figure>
          </section>
          <section anchor="level-1-composed-metrics">
            <name>Level 1 Composed Metrics</name>
            <t>The Metric Type for Level 1 composed metrics is 'level1_composed'.</t>
            <t><strong>Example A: Aggregation-derived (with units)</strong></t>
            <artwork><![CDATA[
Fields:
      Metric_type: level1_composed
      Level: Level 1
      Measurement_Method: hybrid
      Format: unsigned integer
      Length: two octets
      Unit: ms
      Source: aggregation
      Value: 20
]]></artwork>
            <t><strong>Example B: Normalized (unitless)</strong></t>
            <figure anchor="fig-level1-composed-metric">
              <name>Examples of Level 1 composed metrics</name>
              <artwork><![CDATA[
Fields:
      Metric_type: level1_composed
      Level: Level 1
      Measurement_Method: passive
      Format: unsigned integer
      Length: one octet
      Source: normalization
      Value: 8
]]></artwork>
            </figure>
            <t>The <strong>Measurement_Method</strong> of a Level 1 composed metric depends on its contributing inputs. If communication-related metrics obtained via active or hybrid methods are included, the composed metric inherits that method, and if all contributing inputs are collected passively, the method is 'passive'.</t>
          </section>
        </section>
        <section anchor="level-2-representations">
          <name>Level 2 Global Metric</name>
          <t>A Level 2 metric is a single-value, normalized metric that does not
carry any inherent physical unit. While each provider may employ its own
internal methods to compute this value, all providers MUST adhere to the
representation defined in this section to ensure consistent encoding and
interoperable interpretation of the normalized output.</t>
          <t>The Metric Type is 'level2_global' and the Source must be 'normalization'.</t>
          <figure anchor="fig-level-2-metric">
            <name>Example of a normalized Level 2 metric</name>
            <artwork><![CDATA[
Fields:
      Metric_type: level2_global
      Level: Level 2
      Measurement_Method: hybrid
      Format: unsigned integer
      Length: one octet
      Source: normalization
      Value: 1
]]></artwork>
          </figure>
        </section>
      </section>
    </section>
    <section anchor="comparison-among-levels">
      <name>Comparison among Metric Levels</name>
      <t>Metrics are progressively consolidated from Level 0 to Level 1 and then to Level 2, with each level offering an increasing degree of abstraction to address the diverse requirements of different services. Table 1 provides a comparative overview of the defined metric levels.</t>
      <table anchor="comparison">
        <name>Comparison among Metrics Levels</name>
        <thead>
          <tr>
            <th align="center">Level</th>
            <th align="left">Encoding Complexity</th>
            <th align="left">Extensibility</th>
            <th align="left">Stability</th>
            <th align="left">Accuracy</th>
          </tr>
        </thead>
        <tbody>
          <tr>
            <td align="center">Level 0</td>
            <td align="left">High</td>
            <td align="left">Low</td>
            <td align="left">Low</td>
            <td align="left">High</td>
          </tr>
          <tr>
            <td align="center">Level 1</td>
            <td align="left">Medium</td>
            <td align="left">Medium</td>
            <td align="left">Medium</td>
            <td align="left">Medium</td>
          </tr>
          <tr>
            <td align="center">Level 2</td>
            <td align="left">Low</td>
            <td align="left">High</td>
            <td align="left">High</td>
            <td align="left">Low</td>
          </tr>
        </tbody>
      </table>
      <t>Since Level 0 metrics are raw and service-specific, individual services may define their own metric sets, potentially resulting in hundreds or even thousands of distinct metrics across deployments. This diversity introduces significant complexity in protocol encoding and standardization. Consequently, Level 0 metrics are confined to bespoke implementations tailored to specific service needs, rather than being standardized for broad protocol use. In contrast, Level 1 metrics organize raw data into standardized categories, each consolidated into a single value. This structure makes them more suitable for protocol encoding and standardization. The Level 2 metric takes simplification a step further by consolidating all relevant information into a single normalized value, making them the easiest to encode, transmit, and standardize.</t>
      <t>Therefore, from the perspective of encoding complexity, Level 1 and Level 2 metrics are recommended.</t>
      <t>When considering extensibility, Level 0 metrics allow new services to define their own custom metrics. However, this flexibility requires corresponding protocol extensions, and the proliferation of metric types can introduce significant overhead, ultimately reducing the protocol's extensibility. In contrast, Level 1 metrics introduce only a limited set of standardized categories, making protocol extensions more manageable. Level 2 metrics go even further by consolidating all information into a single normalized value, placing the least burden on the protocol.</t>
      <t>Therefore, from an extensibility standpoint, Level 1 and Level 2 metrics are recommended.</t>
      <t>Regarding stability, Level 0 raw metrics would require frequent protocol extensions as new metrics are introduced, leading to an unstable and evolving protocol format. For this reason, standardizing Level 0 metrics within the protocol is not recommended. In contrast, Level 1 metrics involve only a limited set of predefined categories, and Level 2 metrics rely on a single consolidated value, both of which contribute to a more stable and maintainable protocol design.</t>
      <t>Therefore, from a stability standpoint, Level 1 and Level 2 metrics are preferred.</t>
      <t>In conclusion, for CATS, Level 2 metrics are recommended due to their simplicity and minimal protocol overhead. If more advanced scheduling capabilities are required, Level 1 metrics offer a balanced approach with manageable complexity. While Level 0 metrics are the most detailed and dynamic, their high overhead makes them unsuitable for direct transmission to network devices and thus not recommended for standard protocol integration.</t>
    </section>
    <section anchor="cats-metrics-registry">
      <name>CATS Metric Registry Entries</name>
      <t>This section defines the formal registry entries for CATS Level 1 and Level 2 metrics, intended for registration with IANA. The Level 1 registry entries in this section register only the unitless representations of the Level 1 category metrics. As described in <xref target="level-1-representations"/>, Level 1 metrics may be unitless or may retain physical units. Unitless Level 1 metrics may result from normalization or cross-category aggregation. Level 1 metrics that retain physical units (e.g., those derived from spatial or temporal aggregation of Level 0 metrics) are implementation-specific and are not registered by this document.</t>
      <t>Registry entries are separated by measurement methods as classified in <xref target="RFC7799"/>. For Level 1 computing and service metrics, the measurement method is inherited from the underlying data-center OAM and telemetry tools (such as Prometheus <xref target="Prometheus"/> in Kubernetes). Because these metrics are typically collected passively, this document registers only the 'passive' entry for these two categories. For the Level 1 communication metric, 'active', 'passive', and 'hybrid' measurements are supported, and three entries are registered. For the Level 1 composed metric and the Level 2 metric, the measurement method is inherited from the categories of their contributing inputs. When communication-related metrics are included, 'active', 'passive', and 'hybrid' variants apply, so three entries are registered for each of these categories.</t>
      <t>Note that while active probe protocols such as TWAMP <xref target="RFC5357"/> or STAMP <xref target="RFC8762"/> may be extended in future specifications to carry telemetry data from service sites, using a probe packet merely as a transport vehicle does not constitute an active or hybrid measurement of compute or service state per <xref target="RFC7799"/>.</t>
      <t>The detailed measurement procedures for each method are described in the method of measurement subsections of the corresponding registry entries: <xref target="level-2-measurement"/>, <xref target="level-1-computing-measurement"/>, <xref target="level-1-communication-measurement"/>, <xref target="level-1-service-measurement"/>, and <xref target="level-1-composed-measurement"/>.</t>
      <t>By providing a common template that specifies the metric's summary, definition, method of measurement, output, and administrative items, this section ensures interoperability among different implementations.</t>
      <section anchor="cats-level-2-metric-registry">
        <name>CATS Level 2 Metric Registry Entry</name>
        <t>This section gives an initial Registry Entry for the CATS Level 2 metric.</t>
        <section anchor="level-2-summary">
          <name>Summary</name>
          <t>This category includes multiple indexes to the Registry Entry: the element ID, Metric Name, URI, Metric Description, Metric Controller, and Metric Version.</t>
          <section anchor="id-identifier">
            <name>ID (Identifier)</name>
            <t>IANA has allocated the Identifier TBD_1, TBD_2, and TBD_3 for the Named Metric Entries in this section. See the next Section for mapping to Names.</t>
          </section>
          <section anchor="name">
            <name>Name</name>
            <t>TBD_1: Norm_Active_CATS-Level 2_RFCXXXXsecY_Unitless_Singleton</t>
            <t>TBD_2: Norm_Passive_CATS-Level 2_RFCXXXXsecY_Unitless_Singleton</t>
            <t>TBD_3: Norm_Hybrid_CATS-Level 2_RFCXXXXsecY_Unitless_Singleton</t>
            <t>Naming Rule Explanation</t>
            <ul spacing="normal">
              <li>
                <t>Norm: Metric type (Normalized Score)</t>
              </li>
              <li>
                <t>Active, Passive, or Hybrid: Measurement method</t>
              </li>
              <li>
                <t>CATS-Level 2: Metric level (CATS Metric Framework Level 2)</t>
              </li>
              <li>
                <t>RFCXXXXsecY: Specification reference (To-be-assigned RFC number and section number)</t>
              </li>
              <li>
                <t>Unitless: Metric has no units</t>
              </li>
              <li>
                <t>Singleton: Metric is a single value</t>
              </li>
            </ul>
          </section>
          <section anchor="uri">
            <name>URI</name>
            <t>To-be-assigned.</t>
          </section>
          <section anchor="description">
            <name>Description</name>
            <t>This metric represents a single normalized score used within CATS (Level 2). It is derived by aggregating one or more CATS Level 0 and/or Level 1 metrics, followed by a normalization process that produces a unitless value. The resulting score provides a concise assessment of the overall capability of a service instance, enabling rapid comparison across instances and supporting efficient traffic steering decisions.</t>
          </section>
          <section anchor="change-controller">
            <name>Change Controller</name>
            <t>IETF</t>
          </section>
          <section anchor="version">
            <name>Version</name>
            <t>1.0</t>
          </section>
        </section>
        <section anchor="level-2-definition">
          <name>Metric Definition</name>
          <section anchor="reference-definition">
            <name>Reference Definition</name>
            <t>Referenced sections of this document: <xref target="level-2-metric"/> on Level 2 metric definition and <xref target="aggregation-normalization-functions"/> on aggregation and normalization functions.</t>
          </section>
          <section anchor="fixed-parameters">
            <name>Fixed Parameters</name>
            <ul spacing="normal">
              <li>
                <t>Normalization score range: 0-10 (0 indicates the poorest capability, 10 indicates the optimal capability)</t>
              </li>
              <li>
                <t>Data precision: non-negative integer</t>
              </li>
            </ul>
          </section>
        </section>
        <section anchor="level-2-measurement">
          <name>Method of Measurement</name>
          <t>This category includes columns for references to relevant sections of the RFC(s) and any supplemental information needed to ensure an unambiguous method for implementations.</t>
          <section anchor="reference-methods">
            <name>Reference Methods</name>
            <t>Passive measurement: Collect Level 0 service and compute raw metrics using platform-specific management protocols or tools (e.g., Prometheus <xref target="Prometheus"/> in Kubernetes). Collect Level 0 network performance raw metrics using existing standardized protocols (e.g., NETCONF <xref target="RFC6241"/>, IPFIX <xref target="RFC7011"/>). Aggregate and normalize the collected metrics to generate Level 1 category metrics and derive the Level 2 singleton score. No active probe is used.</t>
            <t>Active measurement: Applicable when communication-related inputs are obtained strictly via active measurement tools (e.g., TWAMP <xref target="RFC5357"/> or STAMP <xref target="RFC8762"/>) and combined with site metrics according to operator policy, where the active probing characterizes the path performance component.</t>
            <t>Hybrid measurement: Applicable when active communication measurements are combined with passively collected communication, compute, or service metrics to derive the Level 2 singleton score. (Note: Any protocol extensions required to carry compute or service telemetry inside TWAMP/STAMP packets are out of scope of this document and are left for future specification.)</t>
            <t>Aggregation logic: Refer to <xref target="aggregation-function"/>.</t>
            <t>Normalization logic: Refer to <xref target="normalization-function"/>.</t>
          </section>
          <section anchor="packet-stream-generation">
            <name>Packet Stream Generation</name>
            <t>Passive measurement: N/A.</t>
            <t>Active and hybrid measurement: For the communication measurement component, the C-PS generates probe packets using TWAMP <xref target="RFC5357"/> or STAMP <xref target="RFC8762"/>, and the C-NMA acts as the measurement endpoint or reflector.</t>
          </section>
          <section anchor="traffic-filtering-observation-details">
            <name>Traffic Filtering (Observation) Details</name>
            <t>N/A</t>
          </section>
          <section anchor="sampling-distribution">
            <name>Sampling Distribution</name>
            <t>Passive measurement: continuous sampling (e.g., every 10 seconds).</t>
            <t>Active and hybrid measurement: on-demand or periodic active probing, as configured by operator policy.</t>
          </section>
          <section anchor="runtime-parameters-and-data-format">
            <name>Runtime Parameters and Data Format</name>
            <t>CATS Service Contact Instance ID (CSCI-ID): an identifier of CATS service contact instance. According to <xref target="RFC10053"/>, a unicast IP address can be an example of identifier. (format: ipv4-address-no-zone or ipv6-address-no-zone, complying with <xref target="RFC9911"/>)</t>
            <t>Service_Instance_IP: Service instance IP address (format: ipv4-address-no-zone or ipv6-address-no-zone, complying with <xref target="RFC9911"/>)</t>
            <t>Measurement_Window: Metric measurement time window (Units: seconds, milliseconds; Format: uint64; Default: 10 seconds)</t>
          </section>
          <section anchor="roles">
            <name>Roles</name>
            <t>C-SMA: Collects Level 0 service and compute raw metrics passively via local OAM telemetry tools, and optionally calculates Level 1 compute and service scores.</t>
            <t>C-NMA: Collects Level 0 network performance raw metrics (passively, actively, or via hybrid methods), and optionally calculates Level 1 communication scores.</t>
            <t>C-PS: Aggregates Level 1 metrics (or Level 0 metrics) collected from C-NMA and C-SMA to calculate the Level 2 metric. In active or hybrid modes, it may initiate network probes toward C-NMA to obtain communication metrics.</t>
          </section>
        </section>
        <section anchor="level-2-output">
          <name>Output</name>
          <t>This category specifies all details of the output of measurements using the metric.</t>
          <section anchor="type">
            <name>Type</name>
            <t>Singleton value</t>
          </section>
          <section anchor="reference-definition-1">
            <name>Reference Definition</name>
            <t>Output format: Refer to <xref target="level-2-representations"/> of this document.</t>
            <t>Score semantics: 0-3 (Low capability, not recommended for steering), 4-7 (Medium capability, optional for steering), 8-10 (High capability, priority for steering)</t>
          </section>
          <section anchor="metric-units">
            <name>Metric Units</name>
            <t>Unitless</t>
          </section>
          <section anchor="calibration">
            <name>Calibration</name>
            <t>Calibration method: Conduct benchmark calibration based on standard test sets (fixed workload) to ensure the output score deviation of C-SMA and C-NMA is lower than 0.1 (one abnormal score in every ten test rounds).</t>
          </section>
        </section>
        <section anchor="level-2-administrative-items">
          <name>Administrative Items</name>
          <section anchor="status">
            <name>Status</name>
            <t>Current</t>
          </section>
          <section anchor="requester">
            <name>Requester</name>
            <t>To-be-assgined</t>
          </section>
          <section anchor="revision">
            <name>Revision</name>
            <t>1.0</t>
          </section>
          <section anchor="revision-date">
            <name>Revision Date</name>
            <t>2026-01-20</t>
          </section>
          <section anchor="comments-and-remarks">
            <name>Comments and Remarks</name>
            <t>None</t>
          </section>
        </section>
      </section>
      <section anchor="cats-level-1-computing-metric">
        <name>CATS Level 1 Metric Registry Entry: Computing</name>
        <t>This section gives initial Registry Entries for the CATS Level 1 metric in the <em>computing</em> category.</t>
        <section anchor="level-1-computing-summary">
          <name>Summary</name>
          <t>This category includes multiple indexes to the Registry Entry: the element ID, Metric Name, URI, Metric Description, Metric Controller, and Metric Version.</t>
          <section anchor="id-identifier-1">
            <name>ID (Identifier)</name>
            <t>IANA has allocated the Identifier TBD_4 for the Named Metric Entries in this section. See the next Section for mapping to Names.</t>
          </section>
          <section anchor="name-1">
            <name>Name</name>
            <t>TBD_4: Comb_Passive_CATS-Level 1_Computing_RFCXXXXsecY_Unitless_Singleton</t>
            <t>Naming Rule Explanation</t>
            <ul spacing="normal">
              <li>
                <t>Comb: Metric type (Combined Score)</t>
              </li>
              <li>
                <t>Passive: Measurement method</t>
              </li>
              <li>
                <t>CATS-Level 1: Metric level (CATS Metric Framework Level 1)</t>
              </li>
              <li>
                <t>Computing: Metric category (Computing)</t>
              </li>
              <li>
                <t>RFCXXXXsecY: Specification reference (To-be-assigned RFC number and section number)</t>
              </li>
              <li>
                <t>Unitless: Metric has no units</t>
              </li>
              <li>
                <t>Singleton: Metric is a single value for the computing category</t>
              </li>
            </ul>
          </section>
          <section anchor="uri-1">
            <name>URI</name>
            <t>To-be-assigned.</t>
          </section>
          <section anchor="description-1">
            <name>Description</name>
            <t>This metric represents a single normalized score for the <em>computing</em> category within CATS (Level 1). It is derived from one or more computing-related Level 0 metrics (e.g., CPU/GPU/NPU utilization, CPU frequency, memory utilization, or other computing resource indicators) by applying an implementation-specific aggregation function over the selected Level 0 computing metrics and then applying a normalization function to produce a unitless score.</t>
            <t>The resulting score provides a concise indication of the relative computing capability (or headroom) of a service contact instance for the purpose of instance selection and traffic steering. Higher values indicate better computing capability according to the provider's normalization strategy.</t>
          </section>
          <section anchor="change-controller-1">
            <name>Change Controller</name>
            <t>IETF</t>
          </section>
          <section anchor="version-1">
            <name>Version</name>
            <t>1.0</t>
          </section>
        </section>
        <section anchor="level-1-computing-definition">
          <name>Metric Definition</name>
          <section anchor="reference-definition-2">
            <name>Reference Definition</name>
            <t>Referenced sections of this document: <xref target="level-1-metrics"/> on Level 1 computing metric definition and <xref target="aggregation-normalization-functions"/> on aggregation and normalization functions.</t>
          </section>
          <section anchor="fixed-parameters-1">
            <name>Fixed Parameters</name>
            <ul spacing="normal">
              <li>
                <t>Normalization score range: 0-10 (0 indicates the poorest computing capability, 10 indicates the optimal computing capability)</t>
              </li>
              <li>
                <t>Data precision: non-negative integer</t>
              </li>
              <li>
                <t>Metric type: "level1_computing"</t>
              </li>
              <li>
                <t>Level: Level 1</t>
              </li>
              <li>
                <t>Metric units: Unitless</t>
              </li>
            </ul>
          </section>
        </section>
        <section anchor="level-1-computing-measurement">
          <name>Method of Measurement</name>
          <t>This category includes columns for references to relevant sections of the RFC(s) and any supplemental information needed to ensure an unambiguous method for implementations.</t>
          <section anchor="reference-methods-1">
            <name>Reference Methods</name>
            <t>Raw Metrics collection: Collect computing-related Level 0 raw metrics (e.g., CPU/GPU/NPU, memory, and relevant platform counters) using platform-specific management protocols or tools (e.g., Prometheus <xref target="Prometheus"/> in Kubernetes or equivalent telemetry systems).</t>
            <t>Passive measurement: The measurement method for Level 1 computing metrics is inherited from the OAM and telemetry tools inside the data center. C-SMA collects Level 0 computing metrics passively using platform telemetry such as Prometheus <xref target="Prometheus"/> in Kubernetes.</t>
            <t>Aggregation logic (within computing category): Refer to <xref target="aggregation-function"/> to combine selected Level 0 computing metrics into a single intermediate value prior to normalization. The selection of Level 0 computing metrics and any weights used are implementation-specific.</t>
            <t>Normalization logic: Refer to <xref target="normalization-function"/> to map the aggregated (or directly selected) computing value into the fixed score range.</t>
            <t>The reference method aggregates and normalizes Level 0 computing metrics to generate a single Level 1 computing score ("level1_computing").</t>
          </section>
          <section anchor="packet-stream-generation-1">
            <name>Packet Stream Generation</name>
            <t>N/A.</t>
          </section>
          <section anchor="traffic-filtering-observation-details-1">
            <name>Traffic Filtering (Observation) Details</name>
            <t>N/A</t>
          </section>
          <section anchor="sampling-distribution-1">
            <name>Sampling Distribution</name>
            <t>Passive measurement: continuous sampling (e.g., every 10 seconds).</t>
          </section>
          <section anchor="runtime-parameters-and-data-format-1">
            <name>Runtime Parameters and Data Format</name>
            <t>CATS Service Contact Instance ID (CSCI-ID): an identifier of CATS service contact instance. According to <xref target="RFC10053"/>, a unicast IP address can be an example of identifier. (format: ipv4-address-no-zone or ipv6-address-no-zone, complying with <xref target="RFC9911"/>)</t>
            <t>Service_Instance_IP: Service instance IP address (format: ipv4-address-no-zone or ipv6-address-no-zone, complying with <xref target="RFC9911"/>)</t>
            <t>Measurement_Window: Metric measurement time window (Units: seconds, milliseconds; Format: uint64; Default: 10 seconds)</t>
          </section>
          <section anchor="roles-1">
            <name>Roles</name>
            <t>C-SMA: Collects Level 0 computing metrics passively via local telemetry tools and calculates the Level 1 computing score.</t>
            <t>C-PS: Receives and uses the Level 1 computing score for steering decisions.</t>
            <t>C-NMA: Not required for this metric.</t>
          </section>
        </section>
        <section anchor="level-1-computing-output">
          <name>Output</name>
          <t>This category specifies all details of the output of measurements using the metric.</t>
          <section anchor="type-1">
            <name>Type</name>
            <t>Singleton value</t>
          </section>
          <section anchor="reference-definition-3">
            <name>Reference Definition</name>
            <t>Output format: Refer to <xref target="level-1-representations"/> of this document.</t>
            <t>Score semantics: 0-3 (Low compute capability, not recommended for steering), 4-7 (Medium compute capability, optional for steering), 8-10 (High compute capability, priority for steering)</t>
          </section>
          <section anchor="metric-units-1">
            <name>Metric Units</name>
            <t>Unitless</t>
          </section>
          <section anchor="calibration-1">
            <name>Calibration</name>
            <t>Calibration method: Conduct benchmark calibration based on representative compute workloads (fixed test workload profiles) to align the mapping from Level 0 computing metrics to the Level 1 score, such that score deviation across measurement agents within the same administrative domain is minimized (e.g., less than 0.1 over repeated test rounds).</t>
          </section>
        </section>
        <section anchor="level-1-computing-administrative-items">
          <name>Administrative Items</name>
          <section anchor="status-1">
            <name>Status</name>
            <t>Current</t>
          </section>
          <section anchor="requester-1">
            <name>Requester</name>
            <t>To-be-assgined</t>
          </section>
          <section anchor="revision-1">
            <name>Revision</name>
            <t>1.0</t>
          </section>
          <section anchor="revision-date-1">
            <name>Revision Date</name>
            <t>2026-01-20</t>
          </section>
          <section anchor="comments-and-remarks-1">
            <name>Comments and Remarks</name>
            <t>None</t>
          </section>
        </section>
      </section>
      <section anchor="cats-level-1-communication-metric">
        <name>CATS Level 1 Metric Registry Entry: Communication</name>
        <t>This section gives an initial Registry Entry for the CATS Level 1 metric in the <em>communication</em> category.</t>
        <section anchor="level-1-communication-summary">
          <name>Summary</name>
          <t>This category includes multiple indexes to the Registry Entry: the element ID, Metric Name, URI, Metric Description, Metric Controller, and Metric Version.</t>
          <section anchor="id-identifier-2">
            <name>ID (Identifier)</name>
            <t>IANA has allocated the Identifier TBD_5, TBD_6, and TBD_7 for the Named Metric Entries in this section. See the next Section for mapping to Names.</t>
          </section>
          <section anchor="name-2">
            <name>Name</name>
            <t>TBD_5: Comb_Active_CATS-Level 1_Communication_RFCXXXXsecY_Unitless_Singleton</t>
            <t>TBD_6: Comb_Passive_CATS-Level 1_Communication_RFCXXXXsecY_Unitless_Singleton</t>
            <t>TBD_7: Comb_Hybrid_CATS-Level 1_Communication_RFCXXXXsecY_Unitless_Singleton</t>
            <t>Naming Rule Explanation</t>
            <ul spacing="normal">
              <li>
                <t>Comb: Metric type (Combined Score)</t>
              </li>
              <li>
                <t>Active, Passive, or Hybrid: Measurement method</t>
              </li>
              <li>
                <t>CATS-Level 1: Metric level (CATS Metric Framework Level 1)</t>
              </li>
              <li>
                <t>Communication: Metric category (Communication)</t>
              </li>
              <li>
                <t>RFCXXXXsecY: Specification reference (To-be-assigned RFC number and section number)</t>
              </li>
              <li>
                <t>Unitless: Metric has no units</t>
              </li>
              <li>
                <t>Singleton: Metric is a single value for the communication category</t>
              </li>
            </ul>
          </section>
          <section anchor="uri-2">
            <name>URI</name>
            <t>To-be-assigned.</t>
          </section>
          <section anchor="description-2">
            <name>Description</name>
            <t>This metric represents a single normalized score for the <em>communication</em> category within CATS (Level 1). It is derived from one or more communication-related Level 0 metrics (e.g., throughput, bandwidth, link utilization, loss, delay, jitter, bytes/packets counters, and other network performance indicators) by applying an implementation-specific aggregation function over the selected Level 0 communication metrics and then applying a normalization function to produce a unitless score.</t>
            <t>The resulting score provides a concise indication of the relative communication capability (or headroom) associated with reaching a service contact instance for the purpose of instance selection and traffic steering. Higher values indicate better communication capability according to the provider's normalization strategy.</t>
          </section>
          <section anchor="change-controller-2">
            <name>Change Controller</name>
            <t>IETF</t>
          </section>
          <section anchor="version-2">
            <name>Version</name>
            <t>1.0</t>
          </section>
        </section>
        <section anchor="level-1-communication-definition">
          <name>Metric Definition</name>
          <section anchor="reference-definition-4">
            <name>Reference Definition</name>
            <t>Referenced sections of this document: <xref target="level-1-metrics"/> on Level 1 communication metric definition and <xref target="aggregation-normalization-functions"/> on aggregation and normalization functions.</t>
          </section>
          <section anchor="fixed-parameters-2">
            <name>Fixed Parameters</name>
            <ul spacing="normal">
              <li>
                <t>Normalization score range: 0-10 (0 indicates the poorest communication capability, 10 indicates the optimal communication capability)</t>
              </li>
              <li>
                <t>Data precision: non-negative integer</t>
              </li>
              <li>
                <t>Metric type: "level1_communication"</t>
              </li>
              <li>
                <t>Level: Level 1</t>
              </li>
              <li>
                <t>Metric units: Unitless</t>
              </li>
            </ul>
          </section>
        </section>
        <section anchor="level-1-communication-measurement">
          <name>Method of Measurement</name>
          <t>This category includes columns for references to relevant sections of the RFC(s) and any supplemental information needed to ensure an unambiguous method for implementations.</t>
          <section anchor="reference-methods-2">
            <name>Reference Methods</name>
            <t>Passive measurement: C-NMA collects Level 0 communication metrics using existing standardized protocols and telemetry systems (e.g., NETCONF <xref target="RFC6241"/>, IPFIX <xref target="RFC7011"/>). The metrics are aggregated and normalized according to <xref target="aggregation-function"/> and <xref target="normalization-function"/> to derive the Level 1 communication score. No active probe is used.</t>
            <t>Active measurement: A C-PS or C-NMA initiates active probes using TWAMP <xref target="RFC5357"/> or STAMP <xref target="RFC8762"/> toward a C-NMA. The C-NMA acts as a measurement endpoint or reflector and returns the measured communication metrics. The Level 1 communication score is derived according to <xref target="aggregation-function"/> and <xref target="normalization-function"/>.</t>
            <t>Hybrid measurement: C-PS or C-NMA initiates probes using TWAMP <xref target="RFC5357"/> or STAMP <xref target="RFC8762"/>, and the active measurement results are combined with passively collected Level 0 communication metrics. The combined metrics are aggregated and normalized to derive the Level 1 communication score.</t>
            <t>Aggregation logic (within communication category): Refer to <xref target="aggregation-function"/> (e.g., Weighted Average Aggregation) to combine selected Level 0 communication metrics into a single intermediate value prior to normalization. The selection of Level 0 communication metrics and any weights used are implementation-specific.</t>
            <t>Normalization logic: Refer to <xref target="normalization-function"/> (e.g., Sigmoid Normalization or Min-max scaling) to map the aggregated (or directly selected) communication value into the fixed score range.</t>
            <t>The reference method aggregates and normalizes Level 0 communication metrics to generate a single Level 1 communication score ("level1_communication"). No cross-category aggregation is performed for this metric (i.e., it does not incorporate compute or service metrics).</t>
          </section>
          <section anchor="packet-stream-generation-2">
            <name>Packet Stream Generation</name>
            <t>Passive measurement: N/A.</t>
            <t>Active and hybrid measurement: The C-PS or C-NMA generates probe packets using TWAMP <xref target="RFC5357"/> or STAMP <xref target="RFC8762"/>. The C-NMA acts as the measurement endpoint or reflector for the active portion.</t>
          </section>
          <section anchor="traffic-filtering-observation-details-2">
            <name>Traffic Filtering (Observation) Details</name>
            <t>N/A</t>
          </section>
          <section anchor="sampling-distribution-2">
            <name>Sampling Distribution</name>
            <t>Passive measurement: continuous sampling (e.g., every 10 seconds).</t>
            <t>Active and hybrid measurement: on-demand or periodic active probing, as configured by operator policy.</t>
          </section>
          <section anchor="runtime-parameters-and-data-format-2">
            <name>Runtime Parameters and Data Format</name>
            <t>CATS Service Contact Instance ID (CSCI-ID): an identifier of CATS service contact instance. According to <xref target="RFC10053"/>, a unicast IP address can be an example of identifier. (format: ipv4-address-no-zone or ipv6-address-no-zone, complying with <xref target="RFC9911"/>)</t>
            <t>Service_Instance_IP: Service instance IP address (format: ipv4-address-no-zone or ipv6-address-no-zone, complying with <xref target="RFC9911"/>)</t>
            <t>Measurement_Window: Metric measurement time window (Units: seconds, milliseconds; Format: uint64; Default: 10 seconds)</t>
          </section>
          <section anchor="roles-2">
            <name>Roles</name>
            <t>C-NMA: In passive mode, collects Level 0 communication metrics and calculates the Level 1 communication score. In active and hybrid modes, acts as a measurement endpoint or reflector and returns the Level 1 communication score.</t>
            <t>C-PS: In passive mode, uses the Level 1 communication score for steering decisions. In active and hybrid modes, initiates probes to obtain the Level 1 communication score.</t>
            <t>C-SMA: Not required for this metric.</t>
          </section>
        </section>
        <section anchor="level-1-commmunication-output">
          <name>Output</name>
          <t>This category specifies all details of the output of measurements using the metric.</t>
          <section anchor="type-2">
            <name>Type</name>
            <t>Singleton value</t>
          </section>
          <section anchor="reference-definition-5">
            <name>Reference Definition</name>
            <t>Output format: Refer to <xref target="level-1-representations"/> of this document.</t>
            <t>Score semantics: 0-3 (Low communication capability, not recommended for steering), 4-7 (Medium communication capability, optional for steering), 8-10 (High communication capability, priority for steering)</t>
          </section>
          <section anchor="metric-units-2">
            <name>Metric Units</name>
            <t>Unitless</t>
          </section>
          <section anchor="calibration-2">
            <name>Calibration</name>
            <t>Calibration method: Conduct benchmark calibration based on representative network test profiles (e.g., fixed traffic mixes and path conditions) to align the mapping from Level 0 communication metrics to the Level 1 score, such that score deviation across measurement agents within the same administrative domain is minimized (e.g., less than 0.1 over repeated test rounds).</t>
          </section>
        </section>
        <section anchor="level-1-communication-administrative-items">
          <name>Administrative Items</name>
          <section anchor="status-2">
            <name>Status</name>
            <t>Current</t>
          </section>
          <section anchor="requester-2">
            <name>Requester</name>
            <t>To-be-assgined</t>
          </section>
          <section anchor="revision-2">
            <name>Revision</name>
            <t>1.0</t>
          </section>
          <section anchor="revision-date-2">
            <name>Revision Date</name>
            <t>2026-01-20</t>
          </section>
          <section anchor="comments-and-remarks-2">
            <name>Comments and Remarks</name>
            <t>None</t>
          </section>
        </section>
      </section>
      <section anchor="cats-level-1-service-metric">
        <name>CATS Level 1 Metric Registry Entry: Service</name>
        <t>This section gives initial Registry Entries for the CATS Level 1 metric in the <em>service</em> category.</t>
        <section anchor="level-1-service-summary">
          <name>Summary</name>
          <t>This category includes multiple indexes to the Registry Entry: the element ID, Metric Name, URI, Metric Description, Metric Controller, and Metric Version.</t>
          <section anchor="id-identifier-3">
            <name>ID (Identifier)</name>
            <t>IANA has allocated the Identifier TBD_8 for the Named Metric Entry in this section. See the next Section for mapping to Names.</t>
          </section>
          <section anchor="name-3">
            <name>Name</name>
            <t>TBD_8: Comb_Passive_CATS-Level 1_Service_RFCXXXXsecY_Unitless_Singleton</t>
            <t>Naming Rule Explanation</t>
            <ul spacing="normal">
              <li>
                <t>Comb: Metric type (Combined Score)</t>
              </li>
              <li>
                <t>Passive: Measurement method</t>
              </li>
              <li>
                <t>CATS-Level 1: Metric level (CATS Metric Framework Level 1)</t>
              </li>
              <li>
                <t>Service: Metric category (Service)</t>
              </li>
              <li>
                <t>RFCXXXXsecY: Specification reference (To-be-assigned RFC number and section number)</t>
              </li>
              <li>
                <t>Unitless: Metric has no units</t>
              </li>
              <li>
                <t>Singleton: Metric is a single value for the service category</t>
              </li>
            </ul>
          </section>
          <section anchor="uri-3">
            <name>URI</name>
            <t>To-be-assigned.</t>
          </section>
          <section anchor="description-3">
            <name>Description</name>
            <t>This metric represents a single normalized score for the <em>service</em> category within CATS (Level 1). It is derived from one or more service-related Level 0 metrics that characterize the health and performance of the service instance itself (e.g., service availability, request success rate, admission/overload indicators, tokens per second and/or requests per second, application-level queue depth, and other service KPIs) by applying an implementation-specific aggregation function over the selected Level 0 service metrics and then applying a normalization function to produce a unitless score.</t>
            <t>The resulting score provides a concise indication of the relative service capability (or headroom) of a service contact instance for the purpose of instance selection and traffic steering. Higher values indicate better service capability according to the provider's normalization strategy.</t>
          </section>
          <section anchor="change-controller-3">
            <name>Change Controller</name>
            <t>IETF</t>
          </section>
          <section anchor="version-3">
            <name>Version</name>
            <t>1.0</t>
          </section>
        </section>
        <section anchor="level-1-service-definition">
          <name>Metric Definition</name>
          <section anchor="reference-definition-6">
            <name>Reference Definition</name>
            <t>Referenced sections of this document: <xref target="level-1-metrics"/> on Level 1 service metric definition and <xref target="aggregation-normalization-functions"/> on aggregation and normalization functions.</t>
          </section>
          <section anchor="fixed-parameters-3">
            <name>Fixed Parameters</name>
            <ul spacing="normal">
              <li>
                <t>Normalization score range: 0-10 (0 indicates the poorest service capability, 10 indicates the optimal service capability)</t>
              </li>
              <li>
                <t>Data precision: non-negative integer</t>
              </li>
              <li>
                <t>Metric type: "level1_service"</t>
              </li>
              <li>
                <t>Level: Level 1</t>
              </li>
              <li>
                <t>Metric units: Unitless</t>
              </li>
            </ul>
          </section>
        </section>
        <section anchor="level-1-service-measurement">
          <name>Method of Measurement</name>
          <t>This category includes columns for references to relevant sections of the RFC(s) and any supplemental information needed to ensure an unambiguous method for implementations.</t>
          <section anchor="reference-methods-3">
            <name>Reference Methods</name>
            <t>Passive measurement: The measurement method for Level 1 service metrics is inherited from the OAM and telemetry tools inside the data center. C-SMA collects Level 0 service metrics (e.g., availability, requests per second, tokens per second) passively using platform telemetry (e.g., Prometheus <xref target="Prometheus"/> in Kubernetes).</t>
            <t>Aggregation logic (within service category): Refer to <xref target="aggregation-function"/> (e.g., Weighted Average Aggregation) to combine selected Level 0 service metrics into a single intermediate value prior to normalization. The selection of Level 0 service metrics, any weights used, and any gating logic (e.g., forcing the score to a low value when the instance is unhealthy) are implementation-specific.</t>
            <t>Normalization logic: Refer to <xref target="normalization-function"/> (e.g., Sigmoid Normalization or Min-max scaling) to map the aggregated (or directly selected) service value into the fixed score range.</t>
            <t>The reference method aggregates and normalizes Level 0 service metrics to generate a single Level 1 service score ("level1_service"). No cross-category aggregation is performed for this metric (i.e., it does not incorporate compute or communication metrics).</t>
          </section>
          <section anchor="packet-stream-generation-3">
            <name>Packet Stream Generation</name>
            <t>N/A</t>
          </section>
          <section anchor="traffic-filtering-observation-details-3">
            <name>Traffic Filtering (Observation) Details</name>
            <t>N/A</t>
          </section>
          <section anchor="sampling-distribution-3">
            <name>Sampling Distribution</name>
            <t>Passive measurement: continuous sampling (e.g., every 10 seconds).</t>
          </section>
          <section anchor="runtime-parameters-and-data-format-3">
            <name>Runtime Parameters and Data Format</name>
            <t>CATS Service Contact Instance ID (CSCI-ID): an identifier of CATS service contact instance. According to <xref target="RFC10053"/>, a unicast IP address can be an example of identifier. (format: ipv4-address-no-zone or ipv6-address-no-zone, complying with <xref target="RFC9911"/>)</t>
            <t>Service_Instance_IP: Service instance IP address (format: ipv4-address-no-zone or ipv6-address-no-zone, complying with <xref target="RFC9911"/>)</t>
            <t>Measurement_Window: Metric measurement time window (Units: seconds, milliseconds; Format: uint64; Default: 10 seconds)</t>
          </section>
          <section anchor="roles-3">
            <name>Roles</name>
            <t>C-SMA: Collects Level 0 service metrics passively via local telemetry tools and calculates the Level 1 service score.</t>
            <t>C-PS: Receives and uses the Level 1 service score for steering decisions.</t>
            <t>C-NMA: Not required for this metric.</t>
          </section>
        </section>
        <section anchor="level-1-service-output">
          <name>Output</name>
          <t>This category specifies all details of the output of measurements using the metric.</t>
          <section anchor="type-3">
            <name>Type</name>
            <t>Singleton value</t>
          </section>
          <section anchor="reference-definition-7">
            <name>Reference Definition</name>
            <t>Output format: Refer to <xref target="level-1-representations"/> of this document.</t>
            <t>Score semantics: 0-3 (Low service capability, not recommended for steering), 4-7 (Medium service capability, optional for steering), 8-10 (High service capability, priority for steering)</t>
          </section>
          <section anchor="metric-units-3">
            <name>Metric Units</name>
            <t>Unitless</t>
          </section>
          <section anchor="calibration-3">
            <name>Calibration</name>
            <t>Calibration method: Conduct benchmark calibration based on representative service workload profiles (fixed request mixes and known-good baselines) to align the mapping from Level 0 service metrics to the Level 1 score, such that score deviation across measurement agents within the same administrative domain is minimized (e.g., less than 0.1 over repeated test rounds). Calibration MAY include failure/overload scenarios (e.g., simulated dependency failures or saturation) to ensure score behavior is consistent with operational intent.</t>
          </section>
        </section>
        <section anchor="level-1-service-administrative-items">
          <name>Administrative Items</name>
          <section anchor="status-3">
            <name>Status</name>
            <t>Current</t>
          </section>
          <section anchor="requester-3">
            <name>Requester</name>
            <t>To-be-assigned</t>
          </section>
          <section anchor="revision-3">
            <name>Revision</name>
            <t>1.0</t>
          </section>
          <section anchor="revision-date-3">
            <name>Revision Date</name>
            <t>2026-01-20</t>
          </section>
          <section anchor="comments-and-remarks-3">
            <name>Comments and Remarks</name>
            <t>None</t>
          </section>
        </section>
      </section>
      <section anchor="cats-level-1-composed-metric">
        <name>CATS Level 1 Metric Registry Entry: Composed</name>
        <t>This section gives initial Registry Entries for the CATS Level 1 metric in the <em>composed</em> category.</t>
        <section anchor="level-1-composed-summary">
          <name>Summary</name>
          <t>This category includes multiple indexes to the Registry Entry: the element ID, Metric Name, URI, Metric Description, Metric Controller, and Metric Version.</t>
          <section anchor="id-identifier-4">
            <name>ID (Identifier)</name>
            <t>IANA has allocated the Identifier TBD_9, TBD_10, and TBD_11 for the Named Metric Entries in this section. See the next Section for mapping to Names.</t>
          </section>
          <section anchor="name-4">
            <name>Name</name>
            <t>TBD_9: Comb_Active_CATS-Level 1_Composed_RFCXXXXsecY_Unitless_Singleton</t>
            <t>TBD_10: Comb_Passive_CATS-Level 1_Composed_RFCXXXXsecY_Unitless_Singleton</t>
            <t>TBD_11: Comb_Hybrid_CATS-Level 1_Composed_RFCXXXXsecY_Unitless_Singleton</t>
            <t>Naming Rule Explanation</t>
            <ul spacing="normal">
              <li>
                <t>Comb: Metric type (Combined Score)</t>
              </li>
              <li>
                <t>Active, Passive, or Hybrid: Measurement method</t>
              </li>
              <li>
                <t>CATS-Level 1: Metric level (CATS Metric Framework Level 1)</t>
              </li>
              <li>
                <t>Composed: Metric category (Composed)</t>
              </li>
              <li>
                <t>RFCXXXXsecY: Specification reference (To-be-assigned RFC number and section number)</t>
              </li>
              <li>
                <t>Unitless: Metric has no units</t>
              </li>
              <li>
                <t>Singleton: Metric is a single value for the composed category</t>
              </li>
            </ul>
          </section>
          <section anchor="uri-4">
            <name>URI</name>
            <t>To-be-assigned.</t>
          </section>
          <section anchor="description-4">
            <name>Description</name>
            <t>This metric represents a single normalized score for the <em>composed</em> category within CATS (Level 1). A composed metric is derived by combining multiple lower-level metrics that may span different categories (e.g., compute, communication, and service) and/or multiple components along the request path.</t>
            <t>Typical examples of composed metrics include (but are not limited to) end-to-end delay, application-level response time, or other synthesized indicators that are computed as a function of multiple contributing factors (e.g., the sum of compute processing delay and network transmission delay along the selected path).</t>
            <t>The composed Level 1 score is obtained by applying an implementation-specific aggregation function over the selected contributing Level 0 metrics (and/or previously computed Level 1 category metrics), followed by a normalization function that yields a unitless score. Higher values indicate better composed capability according to the provider's normalization strategy.</t>
          </section>
          <section anchor="change-controller-4">
            <name>Change Controller</name>
            <t>IETF</t>
          </section>
          <section anchor="version-4">
            <name>Version</name>
            <t>1.0</t>
          </section>
        </section>
        <section anchor="level-1-composed-definition">
          <name>Metric Definition</name>
          <section anchor="reference-definition-8">
            <name>Reference Definition</name>
            <t>Referenced sections of this document: <xref target="level-1-metrics"/> on Level 1 composed metric definition and <xref target="aggregation-normalization-functions"/> on aggregation and normalization functions.</t>
          </section>
          <section anchor="fixed-parameters-4">
            <name>Fixed Parameters</name>
            <ul spacing="normal">
              <li>
                <t>Normalization score range: 0-10 (0 indicates the poorest composed capability, 10 indicates the optimal composed capability)</t>
              </li>
              <li>
                <t>Data precision: non-negative integer</t>
              </li>
              <li>
                <t>Metric type: "level1_composed"</t>
              </li>
              <li>
                <t>Level: Level 1</t>
              </li>
              <li>
                <t>Metric units: Unitless</t>
              </li>
            </ul>
          </section>
        </section>
        <section anchor="level-1-composed-measurement">
          <name>Method of Measurement</name>
          <t>This category includes columns for references to relevant sections of the RFC(s) and any supplemental information needed to ensure an unambiguous method for implementations.</t>
          <section anchor="reference-methods-4">
            <name>Reference Methods</name>
            <t>The measurement method for the Level 1 composed metric is inherited from the categories of its contributing inputs:</t>
            <t>Passive measurement: C-PS aggregates passively collected Level 0 and/or Level 1 metrics (e.g., computing, service, and passively collected communication metrics) and applies normalization to derive the Level 1 composed score. No active probe is used.</t>
            <t>Active measurement: Applicable when the composed metric incorporates communication-related metrics measured strictly via active tools (e.g., TWAMP <xref target="RFC5357"/> or STAMP <xref target="RFC8762"/>) alongside other contributing inputs as configured by operator policy.</t>
            <t>Hybrid measurement: Applicable when the composed metric combines active communication measurements (e.g., via TWAMP <xref target="RFC5357"/> or STAMP <xref target="RFC8762"/>) with passively collected communication, compute, or service metrics to derive the Level 1 composed score.</t>
            <t>Aggregation logic (within composed category): Refer to <xref target="aggregation-function"/> (e.g., Weighted Average Aggregation) to combine selected contributing metrics into a single intermediate value prior to normalization. The aggregation function MAY combine Level 0 metrics directly, and/or MAY take as input one or more Level 1 category metrics (e.g., "level1_computing" and "level1_communication"). The selection of contributing metrics, any weights used, and the composition model (e.g., sum of delays, bottleneck/maximum, or weighted utility) are implementation-specific.</t>
            <t>Normalization logic: Refer to <xref target="normalization-function"/> (e.g., Sigmoid Normalization or Min-max scaling) to map the aggregated composed value into the fixed score range.</t>
          </section>
          <section anchor="packet-stream-generation-4">
            <name>Packet Stream Generation</name>
            <t>Passive measurement: N/A.</t>
            <t>Active and hybrid measurement: For the communication measurement component, the C-PS or C-NMA generates probe packets using TWAMP <xref target="RFC5357"/> or STAMP <xref target="RFC8762"/>, and the remote C-NMA acts as the measurement endpoint or reflector.</t>
          </section>
          <section anchor="traffic-filtering-observation-details-4">
            <name>Traffic Filtering (Observation) Details</name>
            <t>N/A</t>
          </section>
          <section anchor="sampling-distribution-4">
            <name>Sampling Distribution</name>
            <t>Passive measurement: continuous sampling (e.g., every 10 seconds).</t>
            <t>Active and hybrid measurement: on-demand or periodic active probing, as configured by operator policy.</t>
          </section>
          <section anchor="runtime-parameters-and-data-format-4">
            <name>Runtime Parameters and Data Format</name>
            <t>CATS Service Contact Instance ID (CSCI-ID): an identifier of CATS service contact instance. According to <xref target="RFC10053"/>, a unicast IP address can be an example of identifier. (format: ipv4-address-no-zone or ipv6-address-no-zone, complying with <xref target="RFC9911"/>)</t>
            <t>Service_Instance_IP: Service instance IP address (format: ipv4-address-no-zone or ipv6-address-no-zone, complying with <xref target="RFC9911"/>)</t>
            <t>Measurement_Window: Metric measurement time window (Units: seconds, milliseconds; Format: uint64; Default: 10 seconds)</t>
          </section>
          <section anchor="roles-4">
            <name>Roles</name>
            <t>C-SMA: Collects passively contributing compute and service metrics and provides the corresponding Level 1 scores (or Level 0 inputs).</t>
            <t>C-NMA: Collects contributing communication metrics (passively, actively, or via hybrid methods) and provides the corresponding Level 1 scores (or Level 0 inputs).</t>
            <t>C-PS: Computes the Level 1 composed score from the collected contributing metrics. In active or hybrid modes, it may initiate network probes toward C-NMA for the communication component.</t>
          </section>
        </section>
        <section anchor="level-1-composed-output">
          <name>Output</name>
          <t>This category specifies all details of the output of measurements using the metric.</t>
          <section anchor="type-4">
            <name>Type</name>
            <t>Singleton value</t>
          </section>
          <section anchor="reference-definition-9">
            <name>Reference Definition</name>
            <t>Output format: Refer to <xref target="level-1-representations"/> of this document.</t>
            <t>Score semantics: 0-3 (Low composed capability, not recommended for steering), 4-7 (Medium composed capability, optional for steering), 8-10 (High composed capability, priority for steering)</t>
          </section>
          <section anchor="metric-units-4">
            <name>Metric Units</name>
            <t>Unitless</t>
          </section>
          <section anchor="calibration-4">
            <name>Calibration</name>
            <t>Calibration method: Conduct benchmark calibration based on representative end-to-end test profiles (fixed request mixes and controlled network/compute conditions) to align the mapping from contributing metrics to the Level 1 composed score. The calibration goal is to minimize score deviation across measurement agents and computation points within the same administrative domain (e.g., less than 0.1 over repeated test rounds). Calibration MAY include failure and saturation scenarios (e.g., compute overload, network congestion, and dependency failures) to ensure the composed score behavior is consistent with operational intent.</t>
          </section>
        </section>
        <section anchor="level-1-composed-administrative-items">
          <name>Administrative Items</name>
          <section anchor="status-4">
            <name>Status</name>
            <t>Current</t>
          </section>
          <section anchor="requester-4">
            <name>Requester</name>
            <t>To-be-assigned</t>
          </section>
          <section anchor="revision-4">
            <name>Revision</name>
            <t>1.0</t>
          </section>
          <section anchor="revision-date-4">
            <name>Revision Date</name>
            <t>2026-01-20</t>
          </section>
          <section anchor="comments-and-remarks-4">
            <name>Comments and Remarks</name>
            <t>None</t>
          </section>
        </section>
      </section>
    </section>
    <section anchor="ops-considerations">
      <name>Operational Considerations</name>
      <t>This section describes operational aspects related to the deployment and operation of CATS metrics in the network. Since CATS metrics directly influence instance selection and traffic steering decisions, operators should consider the following topics when planning, deploying, and operating CATS systems. Accounting, billing, and SLA export are out of scope of this document.</t>
      <section anchor="negotiation">
        <name>Negotiation of Normalization and Aggregation Functions in Multi-Vendor Environments</name>
        <t>Within a single administrative domain, if CATS components (C-SMA, C-NMA, C-PS) are supplied by different vendors, it is essential to ensure that all vendors have a consistent understanding of CATS metric scores, otherwise, steering decisions may be biased. This is also explicitly required in Section 5.4 of <xref target="RFC10053"/>. This subsection provides operational guidance for negotiating these functions in scenarios where multi-vendor components within the same domain need to interoperate.</t>
        <t>** Key points for normalization function negotiation:</t>
        <ul spacing="normal">
          <li>
            <t>Score range: It is recommended to choose a commonly used range that is easy to read and facilitates subsequent processing, for example, uniformly adopt 0-10.</t>
          </li>
          <li>
            <t>Normalization method type: All parties shall agree on using the same type of method, for example, min-max scaling, and specify its key parameters (e.g., upper/lower bounds, reference mean).</t>
          </li>
        </ul>
        <t>** Key points for aggregation function negotiation: aggregation formula type and weighting coefficients, such as weighted sum, weighted product, harmonic mean, along with their corresponding parameters. The specific form and parameters shall be negotiated and unified among all parties.</t>
        <t>** Direction of comparison: explicitly specify whether a higher score indicates better performance or a lower score indicates better performance. It is recommended to adopt a uniform direction, or to specify each metric category separately and clearly document it.</t>
        <t>All negotiation results described above should be compiled into a formal configuration manifest and synchronised, during the initialisation phase and in an offline manner, to those CATS components that require metrics for decision-making. The manifest should be version-controlled to track changes.</t>
        <t>After the system goes live, each component must assume by default that metric values received from other vendors have been processed according to the agreed functions and are fully comparable, and no runtime dynamic negotiation is required. To maintain this comparability, operational calibration is still required. The administrative-domain operator should run benchmark calibration: at initial deployment, after changes to the agreed functions or configuration manifest, and periodically according to operator policy. Re-calibration should be triggered when score deviation for the same metric type from different vendors exceeds the configured threshold, when the proportion of abnormal scores exceeds the configured limit, or when scores persistently diverge from application QoE. Calibration methods, trigger thresholds, and results should be logged and version-controlled together with the configuration manifest.</t>
      </section>
      <section anchor="update-frq">
        <name>Metric Level Selection and Update Frequency Trade-offs</name>
        <t>As discussed in <xref target="comparison-among-levels"/>, the different levels have trade-offs in encoding complexity, scalability, and stability. Level 1 metrics (compute, communication, service, composed) retain independent information per category, making them suitable for scenarios requiring fine-grained visibility and complex steering policies, but they increase signalling overhead and computational complexity. The Level 2 global score provides a single composite value, simplifying policies and reducing overhead, but it hides the contribution of each category. Operators should choose based on policy complexity and the desired granularity of visibility. The choice of metrics and update frequency affect the control plane, and operators should consider their network scale and policy responsiveness requirements to jointly decide on the level and update parameters.</t>
        <t>In multi-vendor deployments, the negotiation and calibration of normalization and aggregation functions affect both Level 1 and Level 2 metrics. When a Level 1 metric of a given category is to be used across vendors, negotiation and calibration should be performed for that category. Level 1 preserves per-category information, which facilitates per-category review and is generally more controllable and less costly to negotiate than a single Level 2 metric. Level 2 metric should be used only after the aggregation functions of all categories and the global normalization function have been agreed, so as to simplify policy and reduce control-plane overhead. If parameter negotiation cannot be completed, the parties may agree to use a Level 0 metric of a specific category with explicit unit and source for decision-making.</t>
        <t>** Advertisement rate limit: it is recommended to limit per-instance metric updates to no more than once per measurement window (e.g., 10 seconds). When scaling to hundreds or thousands of instances, the total control plane load must be evaluated.</t>
        <t>** Strategies to reduce overhead: when the control plane becomes a bottleneck, operators may:</t>
        <ul spacing="normal">
          <li>
            <t>increase the measurement window (e.g., to 30 or 60 seconds) to reduce update frequency.</t>
          </li>
          <li>
            <t>advertise only Level 2 global scores rather than full Level 1 category metrics.</t>
          </li>
          <li>
            <t>use statistical fields (max, min, mean, cur) to provide summarised information without increasing the number of updates.</t>
          </li>
        </ul>
      </section>
      <section anchor="fault-alarm">
        <name>Metric Collection Failures, Fallback Behavior, and Fault Alarms</name>
        <t>** When metrics are unavailable or fail freshness checks, the following fallback actions are recommended:</t>
        <ul spacing="normal">
          <li>
            <t>Use last known good value: C-PS may continue using the most recent valid value for a limited period (e.g., 2 to 3 measurement windows) while waiting for the metric source to recover.</t>
          </li>
          <li>
            <t>Downgrade or exclude the instance: if the metric source exceeds the configured staleness threshold without recovery, the associated instance should be downgraded or removed from steering decisions until a fresh metric is received.</t>
          </li>
          <li>
            <t>Fallback to network-only steering: as a last resort, affected instances may fall back to traditional network-aware steering (ignoring computing metrics).</t>
          </li>
        </ul>
        <t>It should be noted that CATS Level 1 communication metrics are not equivalent to generic network performance metrics such as IGP metrics <xref target="RFC7471"/> or ALTO cost metrics <xref target="RFC9439"/>, even if they may share the same units. CATS Level 1 communication metrics represent the communication dimension of a service site and are collected by the C-SMA, whereas generic network performance metrics are collected by the C-NMA. When CATS metrics and generic network performance metrics coexist, the C-PS should apply an operator policy that assigns different weights or priorities to these two types of metrics for steering decisions. When falling back to network-only steering, the C-PS should ignore CATS metrics and rely solely on generic network performance metrics.</t>
        <t>** Operators should establish monitoring and alarm mechanisms for the following conditions:</t>
        <ul spacing="normal">
          <li>
            <t>Metric freshness failures: timeouts or invalid signatures as defined in SEC-4 from <xref target="sec-considerations"/> should trigger immediate alarms</t>
          </li>
          <li>
            <t>Component unavailability: session interruptions or publication failures of C-SMA, C-NMA, or C-PS must trigger alarms.</t>
          </li>
          <li>
            <t>Score anomalies: including sudden large drops, stuck values, or significant contradictions between Level 1 category scores and the Level 2 composite score, and these should trigger alarms.</t>
          </li>
        </ul>
        <t>Upon alarm triggering, operators are required to identify root causes, apply temporary mitigation policies, and log the event.</t>
      </section>
      <section anchor="metric-validation">
        <name>Validation of Metric Values</name>
        <t>Operators should periodically validate that normalized or aggregated metric values still reflect actual instance capabilities. Validation goals include confirming consistency between metric values and application performance, detecting implementation or configuration anomalies, and providing input for re-calibration decisions.</t>
        <t>Validation may include:</t>
        <ul spacing="normal">
          <li>
            <t>Correlating metric value trends with observed application QoE/SLA metrics.</t>
          </li>
          <li>
            <t>Comparing metric value distributions and trends against known-good reference instances.</t>
          </li>
          <li>
            <t>Running fixed workloads to check for deviations from expected baselines.</t>
          </li>
        </ul>
        <t>Validation frequency and deviation thresholds are determined by operator policy. Persistent divergence between metric values and application QoE should trigger re-calibration.</t>
      </section>
      <section anchor="management-interop">
        <name>Management Interoperability</name>
        <t>This document does not define a YANG model for CATS metric configuration and management. The specific mapping of normalization and aggregation parameters to the YANG model is defined separately in <xref target="I-D.ietf-cats-data-model"/>.</t>
        <t>To help operators understand the management objects, this document distinguishes the following three categories:</t>
        <ul spacing="normal">
          <li>
            <t>Configuration: normalization bounds, aggregation weights, Measurement_Window, score range, comparison direction, and configuration manifest version.</t>
          </li>
          <li>
            <t>Operational state: current Level 1 and Level 2 scores, Source, Observation_Time, Validity_Interval, and freshness status.</t>
          </li>
          <li>
            <t>Statistics: max, min, mean, cur, and historical trends.</t>
          </li>
        </ul>
        <t>Standard management interfaces (e.g., NETCONF/RESTCONF with YANG) may be used to configure and expose these objects.</t>
      </section>
    </section>
    <section anchor="sec-considerations">
      <name>Security Considerations</name>
      <t>CATS metrics are not merely telemetry data, but are important inputs to the traffic steering selection logic. Similar to routing metrics, incorrect or manipulated metrics can directly affect the underlying forwarding behaviour, potentially leading to service disruption, denial-of-service, or policy violation. CATS metrics are dynamic and sensitive. They may indirectly expose physical locations, health status, or service topology information of CATS service contact instances. Access to such metric would allow attackers to sample or correlate metrics to infer internal service instance states and mount targeted attacks. Therefore, the security properties of CATS metrics - integrity, authenticity, controllability, freshness, and confidentiality, are of critical importance. Adequate measures are required to prevent these metrics are exposed to non authorized entites.</t>
      <t><em>Integrity</em>: ensures that metric values have not been altered by unauthorized entities during collection, distribution, or storage.</t>
      <t>SEC-1: All metric messages MUST be protected with cryptographic integrity mechanisms. Receivers MUST verify the integrity of every metric message before using enclosed metrics for traffic steering decisions. Any detected integrity violation MUST cause the relevant metric message to be discarded and SHOULD trigger an alert absent local policy.</t>
      <t><em>Authenticity</em>: ensures that a metric message genuinely originates from the claimed service contact instance or a trusted publisher.</t>
      <t>SEC-2: Each metric publisher MUST be authenticated using strong cryptographic credentials. Metric messages MUST carry a verifiable proof of origin that binds a metric to the publisher's identity. Receivers MUST reject metric messages that cannot be authenticated, regardless of their content.</t>
      <t><em>Controllability</em>: ensures that only explicitly authorized entities are allowed to publish metrics.</t>
      <t>SEC-3: A CATS system MUST enforce fine-grained control policies that authorize which publishers can report metrics for which services. Unauthorized publication attempts MUST be rejected and logged.</t>
      <t><em>Freshness</em>: ensures that metric values are not stale or replayed. Only sufficiently recent metric values are used for decision-making.</t>
      <t>SEC-4: Each metric message MUST be protected against replay and stale value. Receivers and C-PS MUST enforce freshness checks.</t>
      <t><em>Confidentiality</em>: ensures that metric values are not disclosed to unauthorized entities during transmission.</t>
      <t>SEC-5: Metric messages MUST be encrypted during transmission over any network. Encryption keys MUST be managed securely and rotated periodically.</t>
    </section>
    <section anchor="iana">
      <name>IANA Considerations</name>
      <t>This document defines several CATS metric registry entries. IANA is requested to create a new registry titled "CATS Metrics" under a new "Computing-Aware Traffic Steering (CATS)" registry group.</t>
      <t>The initial entries for this registry are defined in <xref target="cats-metrics-registry"/> as follows:</t>
      <table anchor="iana-cats-metrics-table">
        <name>Initial Contents of the CATS Metrics Registry</name>
        <thead>
          <tr>
            <th align="center">Registry ID</th>
            <th align="left">Metric Name</th>
            <th align="left">Description</th>
            <th align="left">Reference</th>
          </tr>
        </thead>
        <tbody>
          <tr>
            <td align="center">TBD_1</td>
            <td align="left">Norm_Active_CATS-Level 2_RFCXXXXsecY_Unitless_Singleton</td>
            <td align="left">Active variant of the Level 2 normalized score</td>
            <td align="left">
              <xref target="cats-level-2-metric-registry"/></td>
          </tr>
          <tr>
            <td align="center">TBD_2</td>
            <td align="left">Norm_Passive_CATS-Level 2_RFCXXXXsecY_Unitless_Singleton</td>
            <td align="left">Passive variant of the Level 2 normalized score</td>
            <td align="left">
              <xref target="cats-level-2-metric-registry"/></td>
          </tr>
          <tr>
            <td align="center">TBD_3</td>
            <td align="left">Norm_Hybrid_CATS-Level 2_RFCXXXXsecY_Unitless_Singleton</td>
            <td align="left">Hybrid variant of the Level 2 normalized score</td>
            <td align="left">
              <xref target="cats-level-2-metric-registry"/></td>
          </tr>
          <tr>
            <td align="center">TBD_4</td>
            <td align="left">Comb_Passive_CATS-Level 1_Computing_RFCXXXXsecY_Unitless_Singleton</td>
            <td align="left">Passive variant of the Level 1 computing score</td>
            <td align="left">
              <xref target="cats-level-1-computing-metric"/></td>
          </tr>
          <tr>
            <td align="center">TBD_5</td>
            <td align="left">Comb_Active_CATS-Level 1_Communication_RFCXXXXsecY_Unitless_Singleton</td>
            <td align="left">Active variant of the Level 1 communication score</td>
            <td align="left">
              <xref target="cats-level-1-communication-metric"/></td>
          </tr>
          <tr>
            <td align="center">TBD_6</td>
            <td align="left">Comb_Passive_CATS-Level 1_Communication_RFCXXXXsecY_Unitless_Singleton</td>
            <td align="left">Passive variant of the Level 1 communication score</td>
            <td align="left">
              <xref target="cats-level-1-communication-metric"/></td>
          </tr>
          <tr>
            <td align="center">TBD_7</td>
            <td align="left">Comb_Hybrid_CATS-Level 1_Communication_RFCXXXXsecY_Unitless_Singleton</td>
            <td align="left">Hybrid variant of the Level 1 communication score</td>
            <td align="left">
              <xref target="cats-level-1-communication-metric"/></td>
          </tr>
          <tr>
            <td align="center">TBD_8</td>
            <td align="left">Comb_Passive_CATS-Level 1_Service_RFCXXXXsecY_Unitless_Singleton</td>
            <td align="left">Passive variant of the Level 1 service score</td>
            <td align="left">
              <xref target="cats-level-1-service-metric"/></td>
          </tr>
          <tr>
            <td align="center">TBD_9</td>
            <td align="left">Comb_Active_CATS-Level 1_Composed_RFCXXXXsecY_Unitless_Singleton</td>
            <td align="left">Active variant of the Level 1 composed score</td>
            <td align="left">
              <xref target="cats-level-1-composed-metric"/></td>
          </tr>
          <tr>
            <td align="center">TBD_10</td>
            <td align="left">Comb_Passive_CATS-Level 1_Composed_RFCXXXXsecY_Unitless_Singleton</td>
            <td align="left">Passive variant of the Level 1 composed score</td>
            <td align="left">
              <xref target="cats-level-1-composed-metric"/></td>
          </tr>
          <tr>
            <td align="center">TBD_11</td>
            <td align="left">Comb_Hybrid_CATS-Level 1_Composed_RFCXXXXsecY_Unitless_Singleton</td>
            <td align="left">Hybrid variant of the Level 1 composed score</td>
            <td align="left">
              <xref target="cats-level-1-composed-metric"/></td>
          </tr>
        </tbody>
      </table>
      <t>For each entry, IANA is requested to assign a unique Identifier (defined in each subsection) from the registry's assignment pool.</t>
      <t>All metric entries have the following common attributes: Change Controller (IETF), and Version. The naming convention and structure follow the definitions in each respective subsection of <xref target="cats-metrics-registry"/>.</t>
    </section>
  </middle>
  <back>
    <references anchor="sec-combined-references">
      <name>References</name>
      <references anchor="sec-normative-references">
        <name>Normative References</name>
        <reference anchor="RFC2119">
          <front>
            <title>Key words for use in RFCs to Indicate Requirement Levels</title>
            <author fullname="S. Bradner" initials="S." surname="Bradner"/>
            <date month="March" year="1997"/>
            <abstract>
              <t>In many standards track documents several words are used to signify the requirements in the specification. These words are often capitalized. This document defines these words as they should be interpreted in IETF documents. This document specifies an Internet Best Current Practices for the Internet Community, and requests discussion and suggestions for improvements.</t>
            </abstract>
          </front>
          <seriesInfo name="BCP" value="14"/>
          <seriesInfo name="RFC" value="2119"/>
          <seriesInfo name="DOI" value="10.17487/RFC2119"/>
        </reference>
        <reference anchor="RFC3339">
          <front>
            <title>Date and Time on the Internet: Timestamps</title>
            <author fullname="G. Klyne" initials="G." surname="Klyne"/>
            <author fullname="C. Newman" initials="C." surname="Newman"/>
            <date month="July" year="2002"/>
            <abstract>
              <t>This document defines a date and time format for use in Internet protocols that is a profile of the ISO 8601 standard for representation of dates and times using the Gregorian calendar.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="3339"/>
          <seriesInfo name="DOI" value="10.17487/RFC3339"/>
        </reference>
        <reference anchor="RFC5357">
          <front>
            <title>A Two-Way Active Measurement Protocol (TWAMP)</title>
            <author fullname="K. Hedayat" initials="K." surname="Hedayat"/>
            <author fullname="R. Krzanowski" initials="R." surname="Krzanowski"/>
            <author fullname="A. Morton" initials="A." surname="Morton"/>
            <author fullname="K. Yum" initials="K." surname="Yum"/>
            <author fullname="J. Babiarz" initials="J." surname="Babiarz"/>
            <date month="October" year="2008"/>
            <abstract>
              <t>The One-way Active Measurement Protocol (OWAMP), specified in RFC 4656, provides a common protocol for measuring one-way metrics between network devices. OWAMP can be used bi-directionally to measure one-way metrics in both directions between two network elements. However, it does not accommodate round-trip or two-way measurements. This memo specifies a Two-Way Active Measurement Protocol (TWAMP), based on the OWAMP, that adds two-way or round-trip measurement capabilities. The TWAMP measurement architecture is usually comprised of two hosts with specific roles, and this allows for some protocol simplifications, making it an attractive alternative in some circumstances. [STANDARDS-TRACK]</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="5357"/>
          <seriesInfo name="DOI" value="10.17487/RFC5357"/>
        </reference>
        <reference anchor="RFC5644">
          <front>
            <title>IP Performance Metrics (IPPM): Spatial and Multicast</title>
            <author fullname="E. Stephan" initials="E." surname="Stephan"/>
            <author fullname="L. Liang" initials="L." surname="Liang"/>
            <author fullname="A. Morton" initials="A." surname="Morton"/>
            <date month="October" year="2009"/>
            <abstract>
              <t>The IETF has standardized IP Performance Metrics (IPPM) for measuring end-to-end performance between two points. This memo defines two new categories of metrics that extend the coverage to multiple measurement points. It defines spatial metrics for measuring the performance of segments of a source to destination path, and metrics for measuring the performance between a source and many destinations in multiparty communications (e.g., a multicast tree). [STANDARDS-TRACK]</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="5644"/>
          <seriesInfo name="DOI" value="10.17487/RFC5644"/>
        </reference>
        <reference anchor="RFC5835">
          <front>
            <title>Framework for Metric Composition</title>
            <author fullname="A. Morton" initials="A." role="editor" surname="Morton"/>
            <author fullname="S. Van den Berghe" initials="S." role="editor" surname="Van den Berghe"/>
            <date month="April" year="2010"/>
            <abstract>
              <t>This memo describes a detailed framework for composing and aggregating metrics (both in time and in space) originally defined by the IP Performance Metrics (IPPM), RFC 2330, and developed by the IETF. This new framework memo describes the generic composition and aggregation mechanisms. The memo provides a basis for additional documents that implement the framework to define detailed compositions and aggregations of metrics that are useful in practice. This document is not an Internet Standards Track specification; it is published for informational purposes.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="5835"/>
          <seriesInfo name="DOI" value="10.17487/RFC5835"/>
        </reference>
        <reference anchor="RFC6241">
          <front>
            <title>Network Configuration Protocol (NETCONF)</title>
            <author fullname="R. Enns" initials="R." role="editor" surname="Enns"/>
            <author fullname="M. Bjorklund" initials="M." role="editor" surname="Bjorklund"/>
            <author fullname="J. Schoenwaelder" initials="J." role="editor" surname="Schoenwaelder"/>
            <author fullname="A. Bierman" initials="A." role="editor" surname="Bierman"/>
            <date month="June" year="2011"/>
            <abstract>
              <t>The Network Configuration Protocol (NETCONF) defined in this document provides mechanisms to install, manipulate, and delete the configuration of network devices. It uses an Extensible Markup Language (XML)-based data encoding for the configuration data as well as the protocol messages. The NETCONF protocol operations are realized as remote procedure calls (RPCs). This document obsoletes RFC 4741. [STANDARDS-TRACK]</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="6241"/>
          <seriesInfo name="DOI" value="10.17487/RFC6241"/>
        </reference>
        <reference anchor="RFC7011">
          <front>
            <title>Specification of the IP Flow Information Export (IPFIX) Protocol for the Exchange of Flow Information</title>
            <author fullname="B. Claise" initials="B." role="editor" surname="Claise"/>
            <author fullname="B. Trammell" initials="B." role="editor" surname="Trammell"/>
            <author fullname="P. Aitken" initials="P." surname="Aitken"/>
            <date month="September" year="2013"/>
            <abstract>
              <t>This document specifies the IP Flow Information Export (IPFIX) protocol, which serves as a means for transmitting Traffic Flow information over the network. In order to transmit Traffic Flow information from an Exporting Process to a Collecting Process, a common representation of flow data and a standard means of communicating them are required. This document describes how the IPFIX Data and Template Records are carried over a number of transport protocols from an IPFIX Exporting Process to an IPFIX Collecting Process. This document obsoletes RFC 5101.</t>
            </abstract>
          </front>
          <seriesInfo name="STD" value="77"/>
          <seriesInfo name="RFC" value="7011"/>
          <seriesInfo name="DOI" value="10.17487/RFC7011"/>
        </reference>
        <reference anchor="RFC7471">
          <front>
            <title>OSPF Traffic Engineering (TE) Metric Extensions</title>
            <author fullname="S. Giacalone" initials="S." surname="Giacalone"/>
            <author fullname="D. Ward" initials="D." surname="Ward"/>
            <author fullname="J. Drake" initials="J." surname="Drake"/>
            <author fullname="A. Atlas" initials="A." surname="Atlas"/>
            <author fullname="S. Previdi" initials="S." surname="Previdi"/>
            <date month="March" year="2015"/>
            <abstract>
              <t>In certain networks, such as, but not limited to, financial information networks (e.g., stock market data providers), network performance information (e.g., link propagation delay) is becoming critical to data path selection.</t>
              <t>This document describes common extensions to RFC 3630 "Traffic Engineering (TE) Extensions to OSPF Version 2" and RFC 5329 "Traffic Engineering Extensions to OSPF Version 3" to enable network performance information to be distributed in a scalable fashion. The information distributed using OSPF TE Metric Extensions can then be used to make path selection decisions based on network performance.</t>
              <t>Note that this document only covers the mechanisms by which network performance information is distributed. The mechanisms for measuring network performance information or using that information, once distributed, are outside the scope of this document.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="7471"/>
          <seriesInfo name="DOI" value="10.17487/RFC7471"/>
        </reference>
        <reference anchor="RFC7799">
          <front>
            <title>Active and Passive Metrics and Methods (with Hybrid Types In-Between)</title>
            <author fullname="A. Morton" initials="A." surname="Morton"/>
            <date month="May" year="2016"/>
            <abstract>
              <t>This memo provides clear definitions for Active and Passive performance assessment. The construction of Metrics and Methods can be described as either "Active" or "Passive". Some methods may use a subset of both Active and Passive attributes, and we refer to these as "Hybrid Methods". This memo also describes multiple dimensions to help evaluate new methods as they emerge.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="7799"/>
          <seriesInfo name="DOI" value="10.17487/RFC7799"/>
        </reference>
        <reference anchor="RFC8174">
          <front>
            <title>Ambiguity of Uppercase vs Lowercase in RFC 2119 Key Words</title>
            <author fullname="B. Leiba" initials="B." surname="Leiba"/>
            <date month="May" year="2017"/>
            <abstract>
              <t>RFC 2119 specifies common key words that may be used in protocol specifications. This document aims to reduce the ambiguity by clarifying that only UPPERCASE usage of the key words have the defined special meanings.</t>
            </abstract>
          </front>
          <seriesInfo name="BCP" value="14"/>
          <seriesInfo name="RFC" value="8174"/>
          <seriesInfo name="DOI" value="10.17487/RFC8174"/>
        </reference>
        <reference anchor="RFC8762">
          <front>
            <title>Simple Two-Way Active Measurement Protocol</title>
            <author fullname="G. Mirsky" initials="G." surname="Mirsky"/>
            <author fullname="G. Jun" initials="G." surname="Jun"/>
            <author fullname="H. Nydell" initials="H." surname="Nydell"/>
            <author fullname="R. Foote" initials="R." surname="Foote"/>
            <date month="March" year="2020"/>
            <abstract>
              <t>This document describes the Simple Two-way Active Measurement Protocol (STAMP), which enables the measurement of both one-way and round-trip performance metrics, like delay, delay variation, and packet loss.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="8762"/>
          <seriesInfo name="DOI" value="10.17487/RFC8762"/>
        </reference>
        <reference anchor="RFC8911">
          <front>
            <title>Registry for Performance Metrics</title>
            <author fullname="M. Bagnulo" initials="M." surname="Bagnulo"/>
            <author fullname="B. Claise" initials="B." surname="Claise"/>
            <author fullname="P. Eardley" initials="P." surname="Eardley"/>
            <author fullname="A. Morton" initials="A." surname="Morton"/>
            <author fullname="A. Akhter" initials="A." surname="Akhter"/>
            <date month="November" year="2021"/>
            <abstract>
              <t>This document defines the format for the IANA Registry of Performance
Metrics. This document also gives a set of guidelines for Registered
Performance Metric requesters and reviewers.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="8911"/>
          <seriesInfo name="DOI" value="10.17487/RFC8911"/>
        </reference>
        <reference anchor="RFC8912">
          <front>
            <title>Initial Performance Metrics Registry Entries</title>
            <author fullname="A. Morton" initials="A." surname="Morton"/>
            <author fullname="M. Bagnulo" initials="M." surname="Bagnulo"/>
            <author fullname="P. Eardley" initials="P." surname="Eardley"/>
            <author fullname="K. D'Souza" initials="K." surname="D'Souza"/>
            <date month="November" year="2021"/>
            <abstract>
              <t>This memo defines the set of initial entries for the IANA Registry of
Performance Metrics. The set includes UDP Round-Trip Latency and
Loss, Packet Delay Variation, DNS Response Latency and Loss, UDP
Poisson One-Way Delay and Loss, UDP Periodic One-Way Delay and Loss,
ICMP Round-Trip Latency and Loss, and TCP Round-Trip Delay and Loss.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="8912"/>
          <seriesInfo name="DOI" value="10.17487/RFC8912"/>
        </reference>
        <reference anchor="RFC9439">
          <front>
            <title>Application-Layer Traffic Optimization (ALTO) Performance Cost Metrics</title>
            <author fullname="Q. Wu" initials="Q." surname="Wu"/>
            <author fullname="Y. Yang" initials="Y." surname="Yang"/>
            <author fullname="Y. Lee" initials="Y." surname="Lee"/>
            <author fullname="D. Dhody" initials="D." surname="Dhody"/>
            <author fullname="S. Randriamasy" initials="S." surname="Randriamasy"/>
            <author fullname="L. Contreras" initials="L." surname="Contreras"/>
            <date month="August" year="2023"/>
            <abstract>
              <t>The cost metric is a basic concept in Application-Layer Traffic
Optimization (ALTO), and different applications may use different
types of cost metrics. Since the ALTO base protocol (RFC 7285)
defines only a single cost metric (namely, the generic "routingcost"
metric), if an application wants to issue a cost map or an endpoint
cost request in order to identify a resource provider that offers
better performance metrics (e.g., lower delay or loss rate), the base
protocol does not define the cost metric to be used.</t>
              <t>This document addresses this issue by extending the specification to
provide a variety of network performance metrics, including network
delay, delay variation (a.k.a. jitter), packet loss rate, hop count,
and bandwidth.</t>
              <t>There are multiple sources (e.g., estimations based on measurements
or a Service Level Agreement) available for deriving a performance
metric. This document introduces an additional "cost-context" field
to the ALTO "cost-type" field to convey the source of a performance
metric.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="9439"/>
          <seriesInfo name="DOI" value="10.17487/RFC9439"/>
        </reference>
        <reference anchor="RFC9911">
          <front>
            <title>Common YANG Data Types</title>
            <author fullname="J. Schönwälder" initials="J." role="editor" surname="Schönwälder"/>
            <date month="December" year="2025"/>
            <abstract>
              <t>This document defines a collection of common data types to be used with the YANG data modeling language. It includes several new type definitions and obsoletes RFC 6991.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="9911"/>
          <seriesInfo name="DOI" value="10.17487/RFC9911"/>
        </reference>
        <reference anchor="RFC10053">
          <front>
            <title>A Framework for Computing-Aware Traffic Steering (CATS)</title>
            <author fullname="C. Li" initials="C." role="editor" surname="Li"/>
            <author fullname="Z. Du" initials="Z." surname="Du"/>
            <author fullname="M. Boucadair" initials="M." role="editor" surname="Boucadair"/>
            <author fullname="L. M. Contreras" initials="L. M." surname="Contreras"/>
            <author fullname="J. Drake" initials="J." surname="Drake"/>
            <date month="October" year="2026"/>
            <abstract>
              <t>This document describes a framework for Computing-Aware Traffic
 Steering (CATS). Specifically, the document identifies a set of CATS
 functional components, describes their interactions, and provides
 illustrative workflows of the control and data planes. The framework
 covers only the case of a single service provider.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="10053"/>
          <seriesInfo name="DOI" value="10.17487/RFC10053"/>
        </reference>
      </references>
      <references anchor="sec-informative-references">
        <name>Informative References</name>
        <reference anchor="RFC10054">
          <front>
            <title>Computing-Aware Traffic Steering (CATS) Problem Statement, Use Cases, and Requirements</title>
            <author fullname="K. Yao" initials="K." surname="Yao"/>
            <author fullname="L. M. Contreras" initials="L. M." surname="Contreras"/>
            <author fullname="H. Shi" initials="H." surname="Shi"/>
            <author fullname="S. Zhang" initials="S." surname="Zhang"/>
            <author fullname="Q. An" initials="Q." surname="An"/>
            <date month="October" year="2026"/>
            <abstract>
              <t>Distributed computing enhances service response time and energy efficiency by utilizing diverse computing facilities for compute-intensive and delay-sensitive services. To optimize throughput and response time, Computing-Aware Traffic Steering (CATS) selects servers and directs traffic based on compute capabilities and resources, rather than static dispatch or connectivity metrics alone. This document outlines the problem statement and scenarios for CATS within a single domain and drives requirements for the CATS framework.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="10054"/>
          <seriesInfo name="DOI" value="10.17487/RFC10054"/>
        </reference>
        <reference anchor="I-D.ietf-cats-data-model">
          <front>
            <title>Data Model for Computing-Aware Traffic Steering (CATS)</title>
            <author fullname="Huijuan Yao" initials="H." surname="Yao">
              <organization>China Mobile</organization>
            </author>
            <author fullname="Changwang Lin" initials="C." surname="Lin">
              <organization>New H3C Technologies</organization>
            </author>
            <author fullname="Zhenqiang Li" initials="Z." surname="Li">
              <organization>China Mobile</organization>
            </author>
            <author fullname="Quan Xiong" initials="Q." surname="Xiong">
              <organization>ZTE Corporation</organization>
            </author>
            <author fullname="Luis M. Contreras" initials="L. M." surname="Contreras">
              <organization>Telefonica</organization>
            </author>
            <date day="23" month="September" year="2026"/>
            <abstract>
              <t>   This document defines a YANG data model for the management of
   Computing-Aware Traffic Steering (CATS) systems.

              </t>
            </abstract>
          </front>
          <seriesInfo name="Internet-Draft" value="draft-ietf-cats-data-model-00"/>
        </reference>
        <reference anchor="performance-metrics" target="https://www.iana.org/assignments/performance-metrics/performance-metrics.xhtml">
          <front>
            <title>performance-metrics</title>
            <author>
              <organization/>
            </author>
            <date year="2020" month="March" day="19"/>
          </front>
        </reference>
        <reference anchor="DMTF" target="https://www.dmtf.org/">
          <front>
            <title>DMTF</title>
            <author>
              <organization/>
            </author>
            <date year="1998"/>
          </front>
        </reference>
        <reference anchor="Prometheus" target="https://prometheus.io/">
          <front>
            <title>Prometheus</title>
            <author>
              <organization/>
            </author>
            <date year="2012"/>
          </front>
        </reference>
        <reference anchor="Min-max-sigmoid" target="https://doi.org/10.1016/C2013-0-18660-6">
          <front>
            <title>Data Mining: Concepts and Techniques (Fourth Edition)</title>
            <author>
              <organization/>
            </author>
            <date year="2023"/>
          </front>
        </reference>
      </references>
    </references>
    <?line 1530?>

<section anchor="appendix-level-0">
      <name>Level 0 Metric Examples</name>
      <t>Several definitions have been developed within the compute and communication industries, as well as through various standardization efforts---such as those by the <xref target="DMTF"/>---that can serve as Level 0 metrics. This section provides illustrative examples.</t>
      <section anchor="compute-raw-metrics">
        <name>Compute Raw Metrics</name>
        <t>This section uses CPU frequency as an example to illustrate the representation of raw computing metrics. The metric type is labeled as compute_CPU_frequency, with the unit specified in GHz. The format supports floating-point values. The corresponding metric fields are defined as follows:</t>
        <figure anchor="fig-compute-raw-metric">
          <name>An Example for Compute Raw Metrics</name>
          <artwork><![CDATA[
Fields:
      Metric_Type: compute_CPU_frequency
      Level: Level 0
      Measurement_Method: passive
      Format: floating point
      Length: four octets
      Unit: GHz
      Source: nominal
      Value: 2.2
]]></artwork>
        </figure>
      </section>
      <section anchor="communication-raw-metrics">
        <name>Communication Raw Metrics</name>
        <t>This section takes the total transmitted bytes (TxBytes) as an example to show the representation of communication raw metrics. TxBytes are named as "communication type_TxBytes". The unit is Mega Bytes (MB). Format is unsigned integer. It will occupy 4 octets. The source of the metric is "Directly measured" and the statistics is "mean". Example:</t>
        <figure anchor="fig-network-raw-metric">
          <name>An Example for Communication Raw Metrics</name>
          <artwork><![CDATA[
Fields:
      Metric_Type: "communication type_TXBytes"
      Level: Level 0
      Measurement_Method: active
      Format: unsigned integer
      Length: four octets
      Unit: MB
      Source: Directly measured
      Statistics: mean
      Value: 100
]]></artwork>
        </figure>
      </section>
      <section anchor="delay-raw-metrics">
        <name>Delay Raw Metrics</name>
        <t>Delay is a kind of synthesized metric which is influenced by computing, storage access, and network transmission. Usually delay refers to the overal processing duration between the arrival time of a specific service request and the departure time of the corresponding service response. It is named as "delay_raw". The format supports floating point. Its unit is microseconds, and it occupies 4 octets. For example:</t>
        <figure anchor="fig-delay-raw-metric">
          <name>An Example for Delay Raw Metrics</name>
          <artwork><![CDATA[
Fields:
      Metric_Type: "delay_raw"
      Level: Level 0
      Measurement_Method: hybrid
      Format: floating point
      Length: four octets
      Unit: microsecond
      Source: aggregation
      Statistics: max
      Value: 231.5
]]></artwork>
        </figure>
      </section>
    </section>
    <section anchor="appendix-examples-norm-agg">
      <name>Examples on Implementations of Normalization and Aggregation Functions</name>
      <t>This section provides some examples on how to apply normalization functions and aggregation functions to derive Level 1 and Level 2 metric values in CATS system.</t>
      <section anchor="implementations-of-normalization-functions">
        <name>Implementations of Normalization Functions</name>
        <t>This example shows how a C-PS derives a Level 2 metric from Level 1 scores using a sigmoid function. The weights and sigmoid parameters are implementation-specific and should be agreed in multi-vendor deployments as described in <xref target="negotiation"/>.</t>
        <t>Assume the C-PS receives:</t>
        <ul spacing="normal">
          <li>
            <t>'level1_computing': 8</t>
          </li>
          <li>
            <t>'level1_communication': 6</t>
          </li>
          <li>
            <t>'level1_service': 9</t>
          </li>
        </ul>
        <t>Using a weighted sum with weights 0.4, 0.3, and 0.3:</t>
        <t>A = 0.4 * 8 + 0.3 * 6 + 0.3 * 9 = 7.7</t>
        <t>Apply a sigmoid function with midpoint 'mu = 5' and slope 'k = 2':</t>
        <t>N(A) = 10 / (1 + exp(-(A - mu) / k))
= 10 / (1 + exp(-(7.7 - 5) / 2))
= 7.94</t>
        <t>After rounding and clamping to 0-10, the Level 2 metric value is 8.</t>
      </section>
      <section anchor="implementations-of-aggregation-functions">
        <name>Implementations of Aggregation Functions</name>
        <t>The following examples illustrate spatial, temporal, and cross-category aggregation. The functions and parameters are examples only.</t>
        <section anchor="spatial-aggregation-using-arithmetic-mean">
          <name>Spatial Aggregation Using Arithmetic Mean</name>
          <t>A C-SMA collects CPU utilization from four service instances: 40%, 60%, 80%, and 20%. Using the arithmetic mean:</t>
          <t>** Spatial_CPU = (40 + 60 + 80 + 20) / 4 = 50% **</t>
          <t>The result retains the input unit (percent) because the inputs are homogeneous. This value may be used as an implementation-specific aggregation-derived Level 1 computing metric, or as an input to a later normalization function.</t>
        </section>
        <section anchor="temporal-aggregation-using-weighted-average">
          <name>Temporal Aggregation Using Weighted Average</name>
          <t>A C-NMA collects delay samples over three measurement windows: 20 ms, 30 ms, and 40 ms. Using a weighted average with weights 0.2, 0.3, and
0.5:</t>
          <t>** Temporal_delay = 0.2 * 20 + 0.3 * 30 + 0.5 * 40 = 33 ms **</t>
          <t>Alternatively, 'max(20, 30, 40) = 40 ms' may be used for the worst-case temporal aggregation.</t>
        </section>
        <section anchor="cross-category-aggregation-using-weighted-sum">
          <name>Cross-Category Aggregation Using Weighted Sum</name>
          <t>After normalization, a C-PS has a computing score of 8, a communication score of 6, and a service score of 9. Using a weighted sum with weights 0.5, 0.3, and 0.2:</t>
          <t>** Combined_score = 0.5 * 8 + 0.3 * 6 + 0.2 * 9 = 7.6 **</t>
          <t>The result is unitless and can be further normalized to produce a Level 2 global metric. Other functions, such as weighted geometric mean and harmonic mean, may be used according to operator policy.</t>
        </section>
      </section>
    </section>
    <section anchor="acknowledgements">
      <name>Acknowledgements</name>
      <t>Authors would like to give special thanks to Adrian Farrel for detailed review as working group chair.</t>
      <t>Special thanks to Jacqueline McCall for Secdir early review, to Fung Lim for Opsdir early review, to Giuseppe Fioccola for Perfmetrdir early review.</t>
    </section>
    <section anchor="contributors" numbered="false" toc="include" removeInRFC="false">
      <name>Contributors</name>
      <contact initials="M." surname="Boucadair" fullname="Mohamed Boucadair">
        <organization>Orange</organization>
        <address>
          <email>mohamed.boucadair@orange.com</email>
        </address>
      </contact>
      <contact initials="Z." surname="Du" fullname="Zongpeng Du">
        <organization>China Mobile</organization>
        <address>
          <email>duzongpeng@chinamobile.com</email>
        </address>
      </contact>
      <contact initials="H." surname="Shi" fullname="Hang Shi">
        <organization>Huawei</organization>
        <address>
          <email>shihang9@huawei.com</email>
        </address>
      </contact>
      <contact initials="S." surname="Dikshit" fullname="Saumya Dikshit">
        <organization>Aruba Networks, Hewlett Packard Enterprise</organization>
        <address>
          <email>saumya.dikshit@hpe.com</email>
        </address>
      </contact>
      <contact initials="M." surname="Zhu" fullname="Mengfei Zhu">
        <organization>China Mobile</organization>
        <address>
          <email>zhumengfei@cmdi.chinamobile.com</email>
        </address>
      </contact>
    </section>
  </back>
  <!-- ##markdown-source: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-->

</rfc>
