An Empirical Study of I/O Bounded Processes and Threads in Operating System

There has been extensive research work in the field of parallel and distributed computing where jobs are granulized into different processors so that the execution time of a program is reduced. However, this is a computationally intensive strategy that consumes an undesirable amount of computing resources. In this paper we concentrate on performance gain without using such consumption of network resources. We split a process to different segments and run them in time multiplexed fashion for performance gain in a single processor. Experimental results show the performance of splits through threads as well as child processes. We conclude that split using child processes perform better than that of using threads especially for I/O intensive jobs.