Explain the Linux CFS scheduler, nice values, real-time scheduling (SCHED_FIFO/RR), CPU affinity, NUMA awareness, and cgroup CPU bandwidth control.
## Problem
You are tuning a production Linux server that runs a mix of latency-sensitive API servers, batch processing jobs, and containerized microservices. The team is seeing unpredictable latency spikes and wants to understand how the Linux scheduler works so they can make informed tuning decisions. Explain the Linux scheduling subsystem in enough depth that the team can: choose the right scheduling policy for each workload, configure CPU affinity and NUMA placement, and set cgroup CPU bandwidth limits for containers.
Sign up to access the full problem
Design canvas, rubric, hints, and model solutions.