Automatic Parallelization: Predicaments in Computationally Expensive Operations

  • Nahar Prayank Prayank Department of Computer Science, Shri Venkateshwara University, Gajraula Distt- Amroha (U.P), India
  • Bhavsar Dharmesh kumar Bhalchandra Department of Computer Science, Shri Venkateshwara University, Gajraula Distt- Amroha (U.P), India.
  • Vishal Bhatnagar Department of Computer Science, Shri Venkateshwara University, Gajraula Distt- Amroha (U.P), India.
  • Richa Tomer Department of Computer Science, Shri Venkateshwara University, Gajraula Distt- Amroha (U.P), India
  • S.K. Agarwal Department of Computer Science, Shri Venkateshwara University, Gajraula Distt- Amroha (U.P), India.
Keywords: Subroutines, Procedures, Parallelization, Compilers

Abstract

The  applications  developed  for  parallelizing  are  either  single  file  based  programs  or  algorithm  structures.  The commercial  large  scale  application  pose  problems  when  performing  automatic  parallelization  and  needs  to  be addressed and hence thought to be implausible. This paper tries to surface the impediments to be addressed so that the  parallelization  techniques  may  be  applied  to  these  applications.  Benchmark  suite  which  are  specifically developed  to  expose  the  computing  requirement  and  are  well  acclaimed  in  the  industry  have  been  adopted. Benchmarks used are from High Performance Group of the Standard Performance rating Corporation (SPEC), both parallel and serial versions of applications are used.

Automatic parellel serial codes are compared with its parellel variants in this paper. Parellelizing compiler is employed that takes language (formula based) codes and inserts Open Multi-Processing directives around loops determined to be autonomous. Different problems were faced by an automatic parelleizing compiler when dealing with full applications procedures, optimization based on superficial techniques, array input, output and size variations multilanguage hurdles, extreme inclination to library (system defined functions) and endowment accumulations. The result presenting in this paper shall benefit parellelizingcompilers with capibilities for handling large scale science and engineering applications.

Published
2014-12-31