Beyond Uptime: Reliability Engineering for AI Infrastructure

Sreejith Kaimal

This is a sample of the book's content.Buy on Leanpub

Table of Contents

Beyond Uptime : Reliability Engineering for AI Infrastructure

Disclaimer & Limitation of Liability

How AI Infrastructure Actually Works
Operation Slowburn
- How can a system be healthy and still be failing?
The Synchronization Problem
Operation Slowburn : A Forensic Reconstruction
- Incident Premise and Cluster Configuration
- The Forensic Investigation
Specialized Networking Technologies
Lossless Fabric Bottlenecks
The Anatomy of Silent Data Corruption
Systematic RCA - The Failure Graph
Generalizing the Incident :A Practitioner Framework
AI Investigating AI : Towards Autonomous Infrastructure

References

Beyond Uptime: Reliability Engineering for AI Infrastructure/Disclaimer & Limitation of Liability/Specialized Networking Technologies

Specialized Networking Technologies

https://leanpub.com/forensic-ai

Limits of Ordinary Networking

https://leanpub.com/forensic-ai

RDMA, RoCE, and InfiniBand

https://leanpub.com/forensic-ai

A Cautionary Case: Fast Transport Is Not Sufficient Alone

https://leanpub.com/forensic-ai

Comparing the Three Fabrics Along the Dimensions That Matter

https://leanpub.com/forensic-ai

Transport Reliability Semantics: What RDMA Guarantees and What It Does Not

https://leanpub.com/forensic-ai

Previous

Operation Slowburn : A Forensic Reconstruction

Up next

Lossless Fabric Bottlenecks

In this chapter

Specialized Networking Technologies
Limits of Ordinary Networking
RDMA, RoCE, and InfiniBand
A Cautionary Case: Fast Transport Is Not Sufficient Alone
Comparing the Three Fabrics Along the Dimensions That Matter
Transport Reliability Semantics: What RDMA Guarantees and What It Does Not