American Journal of Computer Science and Engineering Survey www.pubicon.co.in Review Article A Survey on Serial and Parallel Optimization Techniques Applicable for Matrix Multiplication Algorithm

Parallel algorithms play an imperative role in the high performance computing environment. Dividing a task into the smaller tasks and assigning them to different processors for parallel execution are the two key concepts to evaluate the performance of parallel algorithms. Performance enhancement is essential in large scientific applications, where dense matrix multiplication algorithm is extensively used. Thus, optimization of this algorithm, both for serial and parallel execution would provide upsurge performance. In this paper, a brief systematic survey on serial and parallel optimization techniques applied on matrix multiplication algorithm is carried out