DTM: A Macro Level Tool and Measure for Message Passing Libraries

Abstract

In clusters containing heterogeneous systems for harnessing the computing and other resources, distributed computing tools are employed. Task is submitted to a tool and the actual execution is carried out on aggregated network resources. Tools take care of scheduling, distributing sub-tasks and gathering results along with synchronization and message exchange requirements. They need initialization and synchronization routines for the submitted task. These tools also provide features like transparency, fault tolerance and load balancing. The aim of tool designers should be to provide quality performance with add on request initialization and feature provision. Initialization routines and special features provision take their own time over core distributed computing. In this paper, a tool DTM is implemented. DTM serves to achieve macro level parallelization. It may also be used for placing other distributed computing tools on a performance index, judging their start-up and performance.