Increased computational performance for vector operations on BLAS-1
Keywords:
Scientific computing, BLAS-1, unroll technique, vector programmingAbstract
The functions library, called Basic Linear Algebra Subprograms (BLAS-1), is considered the programming standard in scientific computing. In this work, we focus on the analysis of various code optimization techniques to increase the computational performance of BLAS-1. In particular, we address a combinational approach to explore possible methods of encoding using unroll technique with di
erent levels of depth, vector data programming with MMX and SSE for Intel processors. Using the main functions of BLAS-1, it was determined numerically a computational increase, expressed in mega-ops, up to 52% compared to the optimized BLAS-1 ATLAS library
Downloads
Downloads
Published
Issue
Section
License
The opinions expressed by the authors do not necessarily reflect the position of the publisher of the publication or of UCLA. The total or partial reproduction of the texts published here is authorized, as long as the complete source and the electronic address of this journal are cited.
The authors fully retain the rights to their works, giving the journal the right to be the first publication where the article is presented. The authors have the right to use their articles for any purpose as long as it is done for non-profit. Authors are recommended to disseminate their articles in the final version, after publication in this journal, in the electronic media of the institutions to which they are affiliated or personal digital media.
