HM - DDD _ Datacenter Diagnostics and Debug - workstream - (2026-07-16)

New to the Open Compute Project? Watch “OCP 101: How to Participate in the Open Compute Project” to learn how the community works — and join our mailing list, linked in the 101 video, to stay involved:    • OCP 101: How to Participate in the Open Co...   This OCP Datacenter Diagnostics and Debug workstream call focused on time synchronization and timestamp requirements for correlating failure events across equipment in modern data centers. The group discussed motivating examples such as shared bus-bar power in racks, memory-semantic accelerator interconnects, unexpected thermal/vibration-induced failures, and building-level telemetry correlation, weighing the cost of high-precision protocols like PTP against simpler NTP. They reached rough consensus to target roughly 10 millisecond accuracy with 100 milliseconds as an acceptable range, noting log timestamp resolution and network latency as practical constraints. Next steps include revisiting the spec document, updating code examples, and preparing a status dump example for upcoming meetings. Chapters: 0:00 Opening and submission logistics 4:32 Status and topic planning 8:24 Event correlation problem statement 14:00 Debug war stories and blast radius 18:48 Time sync cost and accuracy 20:20 Choosing acceptable accuracy range 25:40 Requirement decision and wrap-up