[Cats] Re: Introduction to the status of draft-fu-cats-oam-fw-05 and encourage more discussions

[[e-mail]]

Hi Chuanyang,

Thanks for your review and comments! 

Please see inline with Quan->. Thanks!

Best Regards,
Quan


Original


From: 缪川扬10075421
To: 熊泉00091065;
Cc: [e-mail] <[e-mail]>;[e-mail] <[e-mail]>;
Date: 2026年05月29日 09:18
Subject: Re: [Cats] Introduction to the status of draft-fu-cats-oam-fw-05 and encourage more discussions



Dear Authors，
Thanks very much for the introduction of this CATS OAM document. After brief reviewing it, I noticed the following 2 points:
1. Alignment between Architecture and Terminology:
In Section 5.1, the text explicitly categorizes OAM mechanisms into four layers: Link, Path, Instance, and Service OAM. However, Figure 1 (the layered model diagram) does not label an "Instance OAM" layer; instead, it features "SVC-OAM" and "SI-OAM." This creates a discrepancy between the textual description and the illustration. Would the authors consider updating Figure 1 to explicitly reflect the "Instance OAM" layer, or could you please clarify the exact mapping between the "SI-OAM" component and the "Instance OAM" layer?

Quan->Thanks for your remind! It will be updated in next version, all the names will be revised to "Service OAM" and "Instance OAM" to keep consistency. 

2. Functional Scope of SVC-OAM:
In Section 5.1, the description of Service OAM mentions mechanisms such as "application-layer probing and HTTP-based health checks." Meanwhile, Section 5.2.2 states that the SVC-OAM component includes "joint performance measurement" and "end-to-end total delay (network + service processing)." Is the intent for SVC-OAM to define a brand-new, unified end-to-end measurement protocol, or is it expected to reuse or extend existing protocols (such as TWAMP, OWAMP, or even In-situ OAM) to fulfill these joint measurement requirements? Clarifying this point would be highly valuable for implementers.

Quan->This document just defines the functional OAM requirements for the end-to-end CATS service and do not mean to indicate the protocols implementations. But from my view, it would be better to use the existing protocols and the according extensions.

Thanks in advance for your replying.

Best Regards!
Chuanyang







From: 熊泉
To: [e-mail] <[e-mail]>;[e-mail] <[e-mail]>;
Date: 2026年05月28日 15:37
Subject: [Cats] Introduction to the status of draft-fu-cats-oam-fw-05 and encourage more discussions

-- 
Cats mailing list -- [e-mail]
To unsubscribe send an email to [e-mail]


Dear All,

As suggested by the chairs, I'd like to provide a brief introduction to the status of this document to help everyone to better understand its content and encourage further feedback in the mailing list.

https://datatracker.ietf.org/doc/draft-fu-cats-oam-fw/05/

This document (draft-fu-cats-oam-fw-05) proposes an Operations, Administration, and Maintenance (OAM) framework for Computing-Aware Traffic Steering (CATS) systems. It addresses the limitations of traditional network-centric OAM by introducing a multidimensional approach that integrates both network and computing resource monitoring.

Motivation and Problem Statement:
The framework is designed to prevent "service black-holing," where traffic is directed to service instances that are network-reachable but computationally unresponsive. It enables real-time correlation of network metrics with computing metrics to support accurate steering decisions and ensure services are both reachable and performing as expected.

Key Components:
A four-layer OAM model: Link OAM (physical connectivity), Path OAM (transport paths), Instance OAM (service instance liveness and computing metrics), and Service OAM (end-to-end service delivery)
Two functional components: SI-OAM for service instance monitoring and SVC-OAM for end-to-end service verification
Comprehensive OAM requirements: aligned with the operational and management guidelines in [I-D.draft-opsarea-rfc5706bis]

This document has been presented at IETF #124 and #125. We sincerely appreciate the valuable feedback and suggestions received from Adrian Farrel, Erum Welling, Cheng Li, Peng Liu, Haomian Zheng, and other participants. The authors have revised the draft through multiple versions (00 to 05) to incorporate this feedback. Since the last presentation at IETF 125, no further comments have been raised.

The authors welcome additional review and constructive discussion, and we look forward to your feedback as we consider this draft for potential adoption by the CATS working group.Thanks!

Best Regards,
Quan