Network Working Group S. Randriamasy Internet-Draft Nokia Bell Labs Intended status: Informational L. M. Contreras Expires: 5 September 2024 Telefonica J. Ros-Giralt Qualcomm Europe, Inc. R. Schott Deutsche Telekom 4 March 2024 Joint Exposure of Network and Compute Information for Infrastructure- Aware Service Deployment draft-rcr-opsawg-operational-compute-metrics-03 Abstract Service providers are starting to deploy computing capabilities across the network for hosting applications such as distributed AI workloads, AR/VR, vehicle networks, and IoT, among others. In this network-compute environment, knowing information about the availability and state of the underlying communication and compute resources is necessary to determine both the proper deployment location of the applications and the most suitable servers on which to run them. Further, this information is used by numerous use cases with different interpretations. This document proposes an initial approach towards a common understanding and exposure scheme for metrics reflecting compute and communication capabilities. About This Document This note is to be removed before publishing as an RFC. The latest revision of this draft can be found at https://giralt.github.io/draft-rcr-opsawg-operational-compute- metrics/draft-rcr-opsawg-operational-compute-metrics.html. Status information for this document may be found at https://datatracker.ietf.org/doc/draft-rcr-opsawg-operational- compute-metrics/. Source for this draft and an issue tracker can be found at https://github.com/giralt/draft-rcr-opsawg-operational-compute- metrics. Status of This Memo This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79. Randriamasy, et al. Expires 5 September 2024 [Page 1] Internet-Draft TODO - Abbreviation March 2024 Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet- Drafts is at https://datatracker.ietf.org/drafts/current/. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress." This Internet-Draft will expire on 5 September 2024. Copyright Notice Copyright (c) 2024 IETF Trust and the persons identified as the document authors. All rights reserved. This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/ license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Revised BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Revised BSD License. Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 2. Conventions and Definitions . . . . . . . . . . . . . . . . . 4 3. Problem Space and Needs . . . . . . . . . . . . . . . . . . . 4 4. Use Cases . . . . . . . . . . . . . . . . . . . . . . . . . . 6 4.1. Distributed AI Workloads . . . . . . . . . . . . . . . . 6 4.2. Open Abstraction for Edge Computing . . . . . . . . . . . 8 4.3. Optimized Placement of Microservice Components . . . . . 9 5. Production and Consumption Scenarios of Compute-related Information . . . . . . . . . . . . . . . . . . . . . . . 9 5.1. Producers of Compute-Related Information . . . . . . . . 9 5.2. Consumers of Compute-Related Information . . . . . . . . 10 6. Metrics Selection and Exposure . . . . . . . . . . . . . . . 10 6.1. Edge Resources . . . . . . . . . . . . . . . . . . . . . 11 6.2. Network Resources . . . . . . . . . . . . . . . . . . . . 11 6.3. Cloud Resources . . . . . . . . . . . . . . . . . . . . . 11 6.4. Considerations about Metrics . . . . . . . . . . . . . . 12 6.5. Metric Dimensions . . . . . . . . . . . . . . . . . . . . 13 6.6. Abstraction Level and Information Access . . . . . . . . 14 6.7. Distribution and Exposure Mechanisms . . . . . . . . . . 14 Randriamasy, et al. Expires 5 September 2024 [Page 2] Internet-Draft TODO - Abbreviation March 2024 6.7.1. Metric Distribution Computing-Aware Traffic Steering (CATS) . . . . . . . . . . . . . . . . . . . . . . . 15 6.7.2. Metric Exposure with Extensions of ALTO . . . . . . . 15 6.7.3. Exposure of Abstracted Generic Metrics . . . . . . . 15 7. Related Work . . . . . . . . . . . . . . . . . . . . . . . . 16 8. Guiding Principles . . . . . . . . . . . . . . . . . . . . . 17 9. GAP Analysis . . . . . . . . . . . . . . . . . . . . . . . . 17 10. Security Considerations . . . . . . . . . . . . . . . . . . . 18 11. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 18 12. References . . . . . . . . . . . . . . . . . . . . . . . . . 18 12.1. Normative References . . . . . . . . . . . . . . . . . . 18 12.2. Informative References . . . . . . . . . . . . . . . . . 19 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . 20 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 20 1. Introduction Operators are starting to deploy distributed computing environments in different parts of the network with the objective of addressing different service needs including latency, bandwidth, processing capabilities, storage, etc. This translates in the emergence of a number of data centers (both in the cloud and at the edge) of different sizes (e.g., large, medium, small) characterized by distinct dimension of CPUs, memory, and storage capabilities, as well as bandwidth capacity for forwarding the traffic generated in and out of the corresponding data center. The proliferation of the edge computing paradigm further increases the potential footprint and heterogeneity of the environments where a function or application can be deployed, resulting in different unitary cost per CPU, memory, and storage. This increases the complexity of deciding the location where a given function or application should be best deployed or executed. This decision should be jointly influenced on the one hand by the available resources in a given computing environment, and on the other hand by the capabilities of the network path connecting the traffic source with the destination. Network and compute aware function placement and selection has become of utmost importance in the last decade. The availability of such information is taken for granted by the numerous service providers and bodies that are specifying them. However, deployments may reach out to data centers running different implementations with different understandings and representations of compute capabilities and smooth operation is a challenge. While standardization efforts on network capabilities representation and exposure are well-advanced, similar efforts on compute capabilitites are in their infancy. Randriamasy, et al. Expires 5 September 2024 [Page 3] Internet-Draft TODO - Abbreviation March 2024 This document proposes an initial approach towards a common understanding and exposure scheme for metrics reflecting compute capabilities. It aims at leveraging on existing work in the IETF on compute metrics definitions to build synergies. It also aims at reaching out to working or research groups in the IETF that would consume such information and have particular requirements. 2. Conventions and Definitions 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 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as shown here. 3. Problem Space and Needs Visibility and exposure of both (1) network and (2) compute resources to the application is critical to enable the proper functioning of the new class of services arising at the edge (e.g., distributed AI, driverless vehicles, AR/VR, etc.). To understand the problem space and the capabilities that are lacking in today's protocol interfaces needed to enable these new services, we focus on the life cycle of a service. At the edge, compute nodes are deployed near communication nodes (e.g., co-located in a 5G base station) to provide computing services that are close to users with the goal to (1) reduce latency, (2) increase communication bandwidth, (3) enable privacy/personalization (e.g., federated AI learning), and (4) reduce cloud costs and energy. Services are deployed on the communication and compute infrastructure through a two-phase life cycle that involves first a service _deployment stage_ and then a _service selection_ stage (Figure 1). +-------------+ +--------------+ +-------------+ | | | | | | | New +------> Service +------> Service | | Service | | Deployment | | Selection | | | | | | | +-------------+ +--------------+ +-------------+ Figure 1: Service life cycle. *Service deployment.* This phase is carried out by the service provider, and consists in the deployment of a new service (e.g., a distributed AI training/inference, an XR/AR service, etc.) on the communication and compute infrastructure. The service provider needs to properly size the amount of communication and compute resources Randriamasy, et al. Expires 5 September 2024 [Page 4] Internet-Draft TODO - Abbreviation March 2024 assigned to this new service to meet the expected user demand. The decision on where the service is deployed and how many resources are requested from the infrastructure depends on the levels of QoE that the provider wants to guarantee to the user base. To make a proper deployment decision, the provider must have visibility on the resources available from the infrastructure, including communication resources (e.g., latency and bandwidth) and compute (e.g., CPU, GPU, memory, storage). For instance, to run a Large Language Model (LLM) with 175 billion parameters, a total aggregated memory of 400GB and 8 GPUs are needed. The service provider needs an interface to query the infrastructure, extract the available compute and communication resources, and decide which subset of resources are needed to run the service. *Service selection.* This phase is initiated by the user, through a client application that connects to the deployed service. There are two main decisions that must be performed in the service selection stage: compute node selection and path selection. In the compute node selection step, as the service is generally replicated in N locations (e.g., by leveraging a microservices architecture), the application must decide which of the service replicas it connects to. Similar to the service deployment stage, this decision requires knowledge about communication and compute resources available in each replica. On the other hand, in the path selection decision, the application must decide which path it chooses to connect to the service. This decision depends on the communication properties (e.g., bandwidth and latency) of the available paths. Similar to the service deployment case, the service provider needs an interface to query the infrastructure and extract the available compute and communication resources, with the goal to make informed node and path selection decisions. It is also important to note that, ideally, the node and path selection decisions should be jointly optimized, since in general the best end-to-end performance is achieved by jointly taking into account both decisions. In some cases, however, such decisions may be owned by different players. For instance, in some network environments, the path selection may be decided by the network operator, wheres the node selection may be decided by the application. Even in these cases, it is crucial to have a proper interface (for both the network operator and the service provider) to query the available compute and communication resources from the system. Table 1 summarizes the problem space, the information that needs to be exposed, and the stakeholders that need this information. Randriamasy, et al. Expires 5 September 2024 [Page 5] Internet-Draft TODO - Abbreviation March 2024 +====================+===============+==========================+ | Action to take | Information | Who needs it | | | needed | | +====================+===============+==========================+ | Service placement | Compute and | Service provider | | | communication | | +--------------------+---------------+--------------------------+ | Service selection/ | Compute | Network/service provider | | node selection | | and/or application | +--------------------+---------------+--------------------------+ | Service selection/ | Communication | Network/service and/or | | path selection | | application | +--------------------+---------------+--------------------------+ Table 1: Problem space, needs, and stakeholders. 4. Use Cases 4.1. Distributed AI Workloads Generative AI is a technological feat that opens up many applications such as holding conversations, generating art, developing a research paper, or writing software, among many others. Yet this innovation comes with a high cost in terms of processing and power consumption. While data centers are already running at capacity, it is projected that transitioning current search engine queries to leverage generative AI will increase costs by 10 times compared to traditional search methods [DC-AI-COST]. As (1) computing nodes (CPUs, GPUs, and NPUs) are deployed to build the edge cloud leveraging technologies like 5G and (2) with billions of mobile user devices globally providing a large untapped computational platform, shifting part of the processing from the cloud to the edge becomes a viable and necessary step towards enabling the AI-transition. There are at least four drivers supporting this trend: * Computational and energy savings: Due to savings from not needing large-scale cooling systems and the high performance-per-watt efficiency of the edge devices, some workloads can run at the edge at a lower computational and energy cost [EDGE-ENERGY], especially when considering not only processing but also data transport. * Latency: For applications such as driverless vehicles which require real-time inference at very low latency, running at the edge is necessary. Randriamasy, et al. Expires 5 September 2024 [Page 6] Internet-Draft TODO - Abbreviation March 2024 * Reliability and performance: Peaks in cloud demand for generative AI queries can create large queues and latency, and in some cases even lead to denial of service. In some cases, limited or no connectivity requires running the workloads at the edge. * Privacy, security, and personalization: A "private mode" allows users to strictly utilize on-device (or near-the-device) AI to enter sensitive prompts to chatbots, such as health questions or confidential ideas. These drivers lead to a distributed computational model that is hybrid: Some AI workloads will fully run in the cloud, some will fully run in the edge, and some will run both in the edge and in the cloud. Being able to efficiently run these workloads in this hybrid, distributed, cloud-edge environment is necessary given the aforementioned massive energy and computational costs. To make optimized service and workload placement decisions, information about both the compute and communication resources available in the network is necessary too. Consider as an example a large language model (LLM) used to generate text and hold intelligent conversations. LLMs produce a single token per inference, where a token is almost equivalent to a word. Pipelining and parallelization techniques are used to optimize inference, but this means that a model like GPT-3 could potentially go through all 175 billion parameters that are part of it to generate a single word. To efficiently run these computational-intensive workloads, it is necessary to know the availability of compute resources in the distributed system. Suppose that a user is driving a car while conversing with an AI model. The model can run inference on a variety of compute nodes, ordered from lower to higher compute power as follows: (1) the user's phone, (2) the computer in the car, (3) the 5G edge cloud, and (4) the datacenter cloud. Correspondingly, the system can deploy four different models with different levels of precision and compute requirements. The simplest model with the least parameters can run in the phone, requiring less compute power but yielding lower accuracy. Three other models ordered in increasing value of accuracy and computational complexity can run in the car, the edge, and the cloud. The application can identify the right trade-off between accuracy and computational cost, combined with metrics of communication bandwidth and latency, to make the right decision on which of the four models to use for every inference request. Note that this is similar to the resolution/ bandwidth trade-off commonly found in the image encoding problem, where an image can be encoded and transmitted at different levels of resolution depending on the available bandwidth in the communication channel. In the case of AI inference, however, not only bandwidth is a scarce resource, but also compute. ALTO extensions to support the Randriamasy, et al. Expires 5 September 2024 [Page 7] Internet-Draft TODO - Abbreviation March 2024 exposure of compute resources would allow applications to make optimized decisions on selecting the right computational resource, supporting the efficient execution of hybrid AI workloads. 4.2. Open Abstraction for Edge Computing Modern applications such as AR/VR, V2X, or IoT, require bringing compute closer to the edge in order to meet strict bandwidth, latency, and jitter requirements. While this deployment process resembles the path taken by the main cloud providers (notably, AWS, Facebook, Google and Microsoft) to deploy their large-scale datacenters, the edge presents a key difference: datacenter clouds (both in terms of their infrastructure and the applications run by them) are owned and managed by a single organization, whereas edge clouds involve a complex ecosystem of operators, vendors, and application providers, all striving to provide a quality end-to-end solution to the user. This implies that, while the traditional cloud has been implemented for the most part by using vertically optimized and closed architectures, the edge will necessarily need to rely on a complete ecosystem of carefully designed open standards to enable horizontal interoperability across all the involved parties. This document envisions ALTO playing a role as part of the ecosystem of open standards that are necessary to deploy and operate the edge cloud. As an example, consider a user of an XR application who arrives at his/her home by car. The application runs by leveraging compute capabilities from both the car and the public 5G edge cloud. As the user parks the car, 5G coverage may diminish (due to building interference) making the home local Wi-Fi connectivity a better choice. Further, instead of relying on computational resources from the car and the 5G edge cloud, latency can be reduced by leveraging computing devices (PCs, laptops, tablets) available from the home edge cloud. The application's decision to switch from one domain to another, however, demands knowledge about the compute and communication resources available both in the 5G and the Wi-Fi domains, therefore requiring interoperability across multiple industry standards (for instance, IETF and 3GPP on the public side, and IETF and LF Edge [LF-EDGE] on the private home side). ALTO can be positioned to act as an abstraction layer supporting the exposure of communication and compute information independently of the type of domain the application is currently residing in. Future versions of this document will elaborate further on this use case. Randriamasy, et al. Expires 5 September 2024 [Page 8] Internet-Draft TODO - Abbreviation March 2024 4.3. Optimized Placement of Microservice Components Current applications are transitioning from a monolithic service architecture towards the composition of microservice components, following cloud-native trends. The set of microservices can have associated SLOs which impose constraints not only in terms of required compute resources (CPU, storage, ...) dependent on the compute facilities available, but also in terms of performance indicators such as latency, bandwidth, etc, which impose restrictions in the networking capabilities connecting the computing facilities. Even more complex constrains, such as affinity among certain microservices components could require complex calculations for selecting the most appropriate compute nodes taken into consideration both network and compute information. Thus, service/application orchestrators can benefit from the information exposed by ALTO at the time of deciding the placement of the microservices in the network. 5. Production and Consumption Scenarios of Compute-related Information It is important to understand the scenarios of production and consumption of compute-related information in combination with information related to communication. Leveraging such combination enables the possibility of resource and workload placement optimization, leading to both operational cost reductions to the operator and service provider as well as an improvement on the service level experienced by the end users. 5.1. Producers of Compute-Related Information The information relative to compute (i.e., processing capabilities, memory, and storage capacity) can be structured in two ways. On one hand, the information corresponding to the raw compute resources; on the other hand, the information of resources allocated or utilized by a specific application or service function. The former is typically provided by the management systems enabling the virtualization of the physical resources for a later assignment to the processes running on top. Cloud Managers or Virtual Infrastructure Managers are the entities that manage those resources. These management systems offer APIs to access the available resources in the computing facility. Thus, it can be expected that these APIs can be used for the consumption of such information. Once the raw resources are retrieved from the various compute facilities, it could be possible to generate topological network views of them, as being proposed in [I-D.llc-teas-dc-aware-topo-model]. Randriamasy, et al. Expires 5 September 2024 [Page 9] Internet-Draft TODO - Abbreviation March 2024 Regarding the resources allocated or utilized by a specific application or service function, two situations apply: (1) The total allocation and (2) the allocation per service or application. In the first case, the information can be supplied by the virtualization management systems described before. For the specific per-service allocation, it can be expected that the specific management systems of the service or application is capable to provide the resources being used at run time typically as part of the allocated ones. In this last scenario, it is also reasonable to expect the availability of APIs offering this information, even though they can be specific to the service or application. 5.2. Consumers of Compute-Related Information The consumption of compute-related information is relative to the different phases of the service lifecycle. This means that this information can be consumed in different points of time and for different purposes. The expected consumers can be both external or internal to the network. As external consumers it is possible to consider external application management systems requiring resource availability information for service function placement decision, workload migration in the case of consuming raw resources, or requiring information on the usage of resources for service assurance or service scaling, among others. As internal consumers, it is possible to consider network management entities requiring information on the level of resource utilization for traffic steering (as the Path Selector in [I-D.ldbc-cats-framework]), load balance, or analytics, among others. 6. Metrics Selection and Exposure Regarding metrics exposure one can distinguish the topics of (1) how the metrics are exposed and (2) which kind of metrics need to be exposed. The infrastructure resources can be divided into network and compute related resources. Network based resources can roughly be subdivided according to the network structure into edge, backbone, and cloud resources. This section intends to give a brief outlook regarding these resources for stimulating additional discussion with related work going on in other IETF working groups or standardization bodies. Randriamasy, et al. Expires 5 September 2024 [Page 10] Internet-Draft TODO - Abbreviation March 2024 6.1. Edge Resources Edge resources are referring to latency, bandwidth, compute latency or traffic breakout. 6.2. Network Resources Network resources relate to the traditional network infrastructure. The next table provides an overview of some of the commonly used metrics. +==================+ | Kind of Resource | +==================+ | QoS | +------------------+ | Latency | +------------------+ | Bandwidth | +------------------+ | RTT | +------------------+ | Packet Loss | +------------------+ | Jitter | +------------------+ Table 2: Examples of network resource metrics. 6.3. Cloud Resources The next table provides an example of parameters that could be exposed: Randriamasy, et al. Expires 5 September 2024 [Page 11] Internet-Draft TODO - Abbreviation March 2024 +============+=========+=================================+ | CPU | Compute | Available cpu resources | +============+=========+=================================+ | Memory | Compute | Available memory | +------------+---------+---------------------------------+ | Storage | Storage | Available storage | +------------+---------+---------------------------------+ | Configmaps | Object | Configuration and topology maps | +------------+---------+---------------------------------+ | Secrets | Object | Possible secrets | +------------+---------+---------------------------------+ | Pods | Object | Possible pods | +------------+---------+---------------------------------+ | Jobs | Object | Concurrent jobs | +------------+---------+---------------------------------+ | Services | Object | Concurrent services | +------------+---------+---------------------------------+ Table 3: Examples of cloud resource parameters. 6.4. Considerations about Metrics The metrics considered in this document should be used to support decisions for selection and deployment of services and applications. Further iterations of this document may consider additional life cycle operations such as assurance and relevant metrics. The network netrics listed above are specified in a number of IETF documents such as RFC 9439 [I-D.ietf-alto-performance-metrics], which itself leverages on RFC 7679. The work on compute metrics at the IETF, on the other hand, is in its first stages and merely relates to low-level infrastructure metrics such as in [RFC7666]. However: * decisions for service deployment and selection also involve decisions that require an aggregated view for instance at the service level, Randriamasy, et al. Expires 5 September 2024 [Page 12] Internet-Draft TODO - Abbreviation March 2024 * deciding entities may only have partial access to the compute information and actually do not need to have all the details. A number of public tools and methods to test compute facility performances are made available by cloud service providers or service management businesses, see [UPCLOUD] and [IR] to name a few. However, for the proposed performance metrics, their definition and acquisition method may differ from one provider to the other, making it thus challenging to compare performances across different providers. The latter aspect is particularly problematic for applications running at the edge where a complex ecosystem of operators, vendors, and application providers is involved and calls for a common standardized definition. 6.5. Metric Dimensions Upon exploring existing work, this draft proposes to consider a number of dimensions before identifying the compute metrics needed to take a service operation decision. This list is initial and is to be updated upon further discussion. Dimensions helping to identify needed compute metrics: +===========+==================+=================================+ | Dimension | Definition | Examples | +===========+==================+=================================+ | Decision | what are the | monitoring, benchmarking, | | | metrics used for | service selection and placement | +-----------+------------------+---------------------------------+ | Driving | what is assessed | speed, scalability, cost, | | KPI | with the metrics | stability | +-----------+------------------+---------------------------------+ | Decision | different | infrastructure node/cluster, | | scope | granularities | compute service, end-to-end | | | | application | +-----------+------------------+---------------------------------+ | Receiving | receiving | router, centralized controller, | | entity | metrics | application management | +-----------+------------------+---------------------------------+ | Deciding | computing | router, centralized controller, | | entity | decisions | application management | +-----------+------------------+---------------------------------+ Table 4: Dimensions to consider when idenfitying compute metrics. When metrics are documented according to their life cycle action, it allow for a more reliable interpretation and informed utilization of the metrics. The table below provides some examples: Randriamasy, et al. Expires 5 September 2024 [Page 13] Internet-Draft TODO - Abbreviation March 2024 +====================+=============================================+ | Lifecycle action | Example | +====================+=============================================+ | Acquisition method | telemetry, estimation | +--------------------+---------------------------------------------+ | Value processing | aggregation, abstraction | +--------------------+---------------------------------------------+ | Exposure | in-path distribution, off-path distribution | +--------------------+---------------------------------------------+ Table 5: Metrics documented by life cycle action. 6.6. Abstraction Level and Information Access One important aspect to consider is that receiving entities that need to consume metrics to take selection or placement decisions do not always have access to computing information. In particular, several scenarios to be completed upon further discussions, may need to be considered among which: * the consumer is an ISP that does not own the compute infrastructure or has no access to full information. In this case the compute metrics will likely be estimated * the consumer is an application that has no direct access to full information while the ISP has access to both network and compute information. However the ISP is willing to provide guidance to the application with abstract information. * the consumer has access to full network and compute information and wants to use it for fine-grained decision making e.g. at the node/cluster level * the consumer has access to full information but essentially needs guidance with abstracted information. * the consumer has access to information that is abstracted or detailed depending on the metrics. These scenarios further drive the selection of metrics upon the above mentioned dimensions. 6.7. Distribution and Exposure Mechanisms Integration of network and compute metrics in decisions Randriamasy, et al. Expires 5 September 2024 [Page 14] Internet-Draft TODO - Abbreviation March 2024 6.7.1. Metric Distribution Computing-Aware Traffic Steering (CATS) Other existing work at the IETF CATS WG has explored the collection and distribution of computing metrics in [I-D.ldbc-cats-framework]. They consider three deployment models in their deployment considerations: - distributed among network devices directly, - collected by a centralized control plane, - hybrid where a part of computing metrics are distributed among involved network devices, and others may be collected by a centralized control plane. In the hybrid mode, the draft suggests that some static information (e.g., capabilities information) can be distributed among network devices since they are quite stable. Frequent changing information (e.g., resource utilization) can be collected by a centralized control plane to avoid frequent flooding in the distributed control plane. Beside the required extensions to the routing protocols, the hybrid mode stresses the impact of the dynamicity of the distributed metrics and the need to carefully sort out the metric exposure mode w.r.t. their dynamicity. 6.7.2. Metric Exposure with Extensions of ALTO The ALTO protocol has been difined to expose an abstracted network topology and related path costs in [RFC7285]. Its extension RFC 9240 allows to define entities on which properties can be defined, while [I-D.contreras-alto-service-edge] introduces a proposed entity property that allows to consider an entity as both a network element with network related costs and properties and a element of a data centzer with compute related properties. Such an exposure mechanism is particularly useful for decision making entities which are centralized and located off the network paths. 6.7.3. Exposure of Abstracted Generic Metrics In some cases, whether due to unavailable information details or for the sake of simplicity, a consumer may need reliable but simple guidance to select a service. To this end, abstracted generic metrics may be useful. One can consider a generic metric that can be named “computingcost” and is applied to a contact point to one or more edge servers such as a load balancer, for short an edge server, to reflect the network operator policy and preferences. The metric “computingcost” results from an abstraction method that is hidden from users, similarly to the metric “routingcost” defined in [RFC7285]. For instance, “computingcost” may be higher for an edge server located far away, or in disliked geographical areas, or owned by a provider who does not share information with the Internet Service Provider (ISP) or with Randriamasy, et al. Expires 5 September 2024 [Page 15] Internet-Draft TODO - Abbreviation March 2024 which ISP has a poorer commercial agreement. “computingcost” may also reflect environmental preferences in terms, for instance, of energy source, average consumption vs. local climate, location adequacy vs. climate. One may also consider a generic metric named “computingperf”, applied to an edge server, that reflects its performances, based on measurements or estimations by the ISP or combination thereof. An edge server with a higher “computingperf” value will be preferred. “computingperf” can be based on a vector of one or more metrics reflecting, for instance, responsiveness, reliability of cloud services based on metrics such as latency, packet loss, jitter, time to first and/or last byte, or a single value reflecting a global performance score. 7. Related Work Some existing work has explored compute-related metrics. They can be categorized as follows: * References providing raw compute infrastructure metrics: [I-D.contreras-alto-service-edge] includes references to cloud management solutions (i.e., OpenStack, Kubernetes, etc) which administer the virtualization infrastructure, providing information about raw compute infrastructure metrics. Furthermore, [NFV-TST] describes processor, memory and network interface usage metrics. * References providing compute virtualization metrics: [RFC7666] provides several metrics as part of the Management Information Base (MIB) definition for managing virtual machines controlled by a hypervisor. The objects there defined make reference to the resources consumed by a particluar virtual machine serving as host for services or applications. Moreover, [NFV-INF] provides metrics associated to virtualized network functions. * References providing service metrics including compute-related information: [I-D.dunbar-cats-edge-service-metrics] proposes metrics associated to services running in compute infrastructures. Some of these metrics do not depend on the infrastructure behavior itself but from where such compute infrastructure is topologically located. Other existing work at the IETF CATS WG has explored the collection and distribution of computing metrics in [I-D.ldbc-cats-framework]. In their deployment considerations, they consider three models: distributed, centralized and hybrid. Randriamasy, et al. Expires 5 September 2024 [Page 16] Internet-Draft TODO - Abbreviation March 2024 8. Guiding Principles The driving principles for designing an interface to jointly extract network and compute information are as follows: P1. Leverage metrics across working groups to avoid reinventing the wheel. For instance: * RFC 9439 [I-D.ietf-alto-performance-metrics] leverages IPPM metrics from RFC 7679. * Section 5.2 of [I-D.du-cats-computing-modeling-description] considers delay as a good metric, since it is easy to use in both compute and communication domains. RFC 9439 also defines delay as part of the performance metrics. * Section 6 of [I-D.du-cats-computing-modeling-description] proposes to represent the network structure as graphs, which is similar to the ALTO map services in [RFC7285]. P2. Aim for simplicity, while ensuring the combined efforts don’t leave technical gaps in supporting the full life cycle of service deployment and selection. For instance, the CATS working group is covering path selection from a network standpoint, while ALTO (e.g., [RFC7285]) covers exposing of network information to the service provider and the client application. However, there is currently no effort being pursued to expose compute information to the service provider and the client application for service placement or selection. 9. GAP Analysis From this related work it is evident that compute-related metrics can serve several purposes, ranging from service instance instantiation to service instance behavior, and then to service instance selection. Some of the metrics could refer to the same object (e.g., CPU) but with a particular usage and scope. In contrast, the network metrics are more uniform and straightforward. It is then necessary to consistently define a set of metrics that could assist to the operation in the different concerns identified so far, so that networks and systems could have a common understanding of the perceived compute performance. When combined with network metrics, the combined network plus compute performance behavior will assist informed decisions particular to each of the operational concerns related to the different parts of a service life cycle. Randriamasy, et al. Expires 5 September 2024 [Page 17] Internet-Draft TODO - Abbreviation March 2024 10. Security Considerations TODO Security 11. IANA Considerations This document has no IANA actions. 12. References 12.1. Normative References [I-D.du-cats-computing-modeling-description] Du, Z., Fu, Y., Li, C., Huang, D., and Z. Fu, "Computing Information Description in Computing-Aware Traffic Steering", Work in Progress, Internet-Draft, draft-du- cats-computing-modeling-description-02, 23 October 2023, . [I-D.ietf-alto-performance-metrics] Wu, Q., Yang, Y. R., Lee, Y., Dhody, D., Randriamasy, S., and L. M. Contreras, "Application-Layer Traffic Optimization (ALTO) Performance Cost Metrics", Work in Progress, Internet-Draft, draft-ietf-alto-performance- metrics-28, 21 March 2022, . [I-D.ldbc-cats-framework] Li, C., Du, Z., Boucadair, M., Contreras, L. M., and J. Drake, "A Framework for Computing-Aware Traffic Steering (CATS)", Work in Progress, Internet-Draft, draft-ldbc- cats-framework-06, 8 February 2024, . [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/RFC2119, March 1997, . [RFC7285] Alimi, R., Ed., Penno, R., Ed., Yang, Y., Ed., Kiesel, S., Previdi, S., Roome, W., Shalunov, S., and R. Woundy, "Application-Layer Traffic Optimization (ALTO) Protocol", RFC 7285, DOI 10.17487/RFC7285, September 2014, . Randriamasy, et al. Expires 5 September 2024 [Page 18] Internet-Draft TODO - Abbreviation March 2024 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, May 2017, . 12.2. Informative References [DC-AI-COST] "Generative AI Breaks The Data Center - Data Center Infrastructure And Operating Costs Projected To Increase To Over $76 Billion By 2028", Forbes, Tirias Research Report , 2023. [EDGE-ENERGY] "Estimating energy consumption of cloud, fog, and edge computing infrastructures", IEEE Transactions on Sustainable Computing , 2019. [I-D.contreras-alto-service-edge] Contreras, L. M., Randriamasy, S., Ros-Giralt, J., Perez, D. A. L., and C. E. Rothenberg, "Use of ALTO for Determining Service Edge", Work in Progress, Internet- Draft, draft-contreras-alto-service-edge-10, 13 October 2023, . [I-D.dunbar-cats-edge-service-metrics] Dunbar, L., Majumdar, K., Mishra, G. S., Wang, H., and H. Song, "5G Edge Services Use Cases", Work in Progress, Internet-Draft, draft-dunbar-cats-edge-service-metrics-01, 6 July 2023, . [I-D.llc-teas-dc-aware-topo-model] Lee, Y., Liu, X., and L. M. Contreras, "DC aware TE topology model", Work in Progress, Internet-Draft, draft- llc-teas-dc-aware-topo-model-03, 10 July 2023, . [IR] "Cloud Performance Testing Best Tips and Tricks", n.d., . [LF-EDGE] "Linux Foundation Edge", https://www.lfedge.org/ , March 2023. [NFV-INF] "ETSI GS NFV-INF 010, v1.1.1, Service Quality Metrics", 1 December 2014, . Randriamasy, et al. Expires 5 September 2024 [Page 19] Internet-Draft TODO - Abbreviation March 2024 [NFV-TST] "ETSI GS NFV-TST 008 V3.3.1, NFVI Compute and Network Metrics Specification", 1 June 2020, . [RFC7666] Asai, H., MacFaden, M., Schoenwaelder, J., Shima, K., and T. Tsou, "Management Information Base for Virtual Machines Controlled by a Hypervisor", RFC 7666, DOI 10.17487/RFC7666, October 2015, . [UPCLOUD] "How to benchmark Cloud Servers", May 2023, . Acknowledgments TODO acknowledge. Authors' Addresses S. Randriamasy Nokia Bell Labs Email: sabine.randriamasy@nokia-bell-labs.com L. M. Contreras Telefonica Email: luismiguel.contrerasmurillo@telefonica.com Jordi Ros-Giralt Qualcomm Europe, Inc. Email: jros@qti.qualcomm.com Roland Schott Deutsche Telekom Email: Roland.Schott@telekom.de Randriamasy, et al. Expires 5 September 2024 [Page 20]