Performance of scheduling strategies for client-server systems

Two approaches to the improvement of the performance of client-server syste ms, multithreading and scheduling of servers, are investigated. Both of the se approaches are observed to have a significant impact on system performan ce. The use of multithreading improves throughput characteristics of system s, whereas the deployment of appropriate scheduling strategies at servers c an produce a significant improvement in mean client response limes. Based o n a simulation model a number of basic questions that are important in the context of scheduling on nonmultithreaded, as well as multithreaded, system s are analyzed. Two important factors, monopolization of servers by large r equests and software bottlenecks, are observed to be important in the conte xt of scheduling on client-server systems. Both server scheduling, as well as multithreading, can be used to control these effects and lead to a highe r system performance. Scheduling policies based on request characteristics are observed to perform well. A new request characteristic that is useful i n the scheduling of client-server systems in the presence of software bottl enecks is proposed. Selection of both the server process, as well as the th read within the server, is required when multiple server, are co-located on the same CPU. A comparison between two scheduling approaches; single level and two level is presented in the paper. The results of this research are useful primarily in the design of operating systems for client-server syste ms and are also of interest to system designers and users. (C) 1999 Academi c Press.