Web servers under overload: How scheduling can help

This paper provides a detailed implementation study of the behavior of web servers serving static requests, where the load fluctuates over time (transient overload). Various external factors are considered including WAN delays and losses, and di#erent client behavior models. We find that performance can be dramatically improved via a kernel-level modification to the web server to change the scheduling policy at the server from the standard FAIR (processor-sharing) scheduling to SRPT (Shortest-Remaining-Processing-Time) scheduling. We find that SRPT scheduling induces no penalties. In particular, throughput is not sacrificed and requests for long files experience only negligibly higher response times under SRPT than they did under the original FAIR scheduling