An End-to-End Assurance Framework for AI/ML Workloads in Datacenters

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gupta, Jit, Banka, Tarun, Gupta, Rahul, Dharmaraj, Mithun, Kaur, Jasleen
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915371604246528
author Gupta, Jit
Banka, Tarun
Gupta, Rahul
Dharmaraj, Mithun
Kaur, Jasleen
author_facet Gupta, Jit
Banka, Tarun
Gupta, Rahul
Dharmaraj, Mithun
Kaur, Jasleen
contents Modern machine learning workloads such as large language model training, fine-tuning jobs are highly distributed and span across hundreds of systems with multiple GPUs. Job completion time for these workloads is the artifact of the application, compute, network and storage performance. In case of failure or degraded performance it is imperative to understand the root cause and possible remediation for the problem for end-to-end assurance. This demo showcases SaaSbased observability and automated troubleshooting for AI/ML workload performance issues using cross-layer telemetry and logs (e.g., Application telemetry, Collective communication logs, GPU Health metrics, Network Flow Data, NIC ROCEv2 telemetry). Different use cases are demonstrated for end-to-end assurance such as Cross-layer Dependency Graph, Cross-layer Service Level Expectations, Automated Root Cause Analysis, GPU-toGPU application path tracing.
format Preprint
id arxiv_https___arxiv_org_abs_2507_03158
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle An End-to-End Assurance Framework for AI/ML Workloads in Datacenters
Gupta, Jit
Banka, Tarun
Gupta, Rahul
Dharmaraj, Mithun
Kaur, Jasleen
Networking and Internet Architecture
Modern machine learning workloads such as large language model training, fine-tuning jobs are highly distributed and span across hundreds of systems with multiple GPUs. Job completion time for these workloads is the artifact of the application, compute, network and storage performance. In case of failure or degraded performance it is imperative to understand the root cause and possible remediation for the problem for end-to-end assurance. This demo showcases SaaSbased observability and automated troubleshooting for AI/ML workload performance issues using cross-layer telemetry and logs (e.g., Application telemetry, Collective communication logs, GPU Health metrics, Network Flow Data, NIC ROCEv2 telemetry). Different use cases are demonstrated for end-to-end assurance such as Cross-layer Dependency Graph, Cross-layer Service Level Expectations, Automated Root Cause Analysis, GPU-toGPU application path tracing.
title An End-to-End Assurance Framework for AI/ML Workloads in Datacenters
topic Networking and Internet Architecture
url https://arxiv.org/abs/2507.03158