A controlled EEVDF vs. CFS comparison
We compared EEVDF and CFS on equivalent 40-CPU CloudLab bare-metal nodes under varied contention and priority.
OPERATING SYSTEMS · SCHEDULING
Linux Scheduler Performance: EEVDF vs. CFS
I benchmarked EEVDF against CFS on 40-CPU bare metal and used eBPF traces to explain why neither scheduler wins every workload.
EEVDF · 4–6× lower p99.9 wakeup latency
EEVDF · up to 65% higher Redis GET throughput
CFS · 13–59% lower HTTP latency
CFS · 10–15% higher 7-Zip throughput
Comparing schedulers through real workloads
Latency-sensitive applications paired with sustained CPU pressure, measured on what users actually notice.
Explaining the results below the application
Application timing showed which scheduler won; eBPF tracing showed why.
The trade-off the comparison revealed
Neither scheduler wins everywhere: EEVDF protects worst-case latency, CFS favors throughput through longer contiguous runs.