Reliability, Availability, and Serviceability

The Reliability, Availability, and Serviceability (RAS) team at IBM’s Linux Technology Center (LTC) contributes broadly to RAS areas of Linux and OPAL firmware. The team works on tracing infrastructure (static and dynamic) for the Linux kernel and user space and its associated tooling such as perf. The primary focus is on IBM’s Power platform and its capabilities, including infrastructure for First Failure Data Capture (FFDC) and tooling to analyze the captured data. On the OPAL front, the team is responsible for making the OPAL-based Power platform robust against hardware errors such as CPU, cache, memory, and clock, provide error logging capabilities and firmware-assisted RAS capabilities for better service of the system.

In keeping with the rich tradition of RAS on Power Systems, the team also maintains an extensive suite of tools exclusive to the platform. More recently, the team has ventured into RAS aspects of cloud and containers — making perf container aware.