The Effect of Scheduling Discipline on CPU-MEM Load Sharing System

Abstract

Scheduling schemes have different effects on the performance of load sharing in distributed systems. Because the load sharing policy based on CPU-memory has made the system memory an important role in effecting the system performance, it can reduce the paging fault and enhance the usage of the system resources, thus can improve the whole performance a lot. In light of the characteristics of the load sharing based on the CPU-memory, a new CPU local scheduling policy named more memory-request more CPU slice (RR-MMMCS), which is based on the round robin policy, is developed in this paper. And the merits of RR-MMMCS have been proved in the trace-driven simulation experiments. Furthermore, we discuss the effects on the load sharing policy based on the CPU-memory of the variance of the interarrival time and service time, and show that the load sharing policy based on the CPU-memory and the RR-MMMCS scheduling policy are effective and have steady performances in both the CPU-bound and the memory-bound jobs.