Loading...

Efficient SIMD numerical interpolation

Ahmadi, H ; Sharif University of Technology | 2005

286 Viewed
  1. Type of Document: Article
  2. DOI: 10.1007/11557654_21
  3. Publisher: Springer Verlag , 2005
  4. Abstract:
  5. This paper reports the results of SIMD implementation of a number of interpolation algorithms on common personal computers. These methods fit a curve on some given input points for which a mathematical function form is not known. We have implemented four widely used methods using vector processing capabilities embedded in Pentium processors. By using SSE (streaming SIMD extension) we could perform all operations on four packed single-precision (32-bit) floating point values simultaneously. Therefore, the running time decreases three times or even more depending on the number of points and the interpolation method. We have implemented four interpolation methods using SSE technology then analyzed their speedup as a function of the number of points being interpolated. A comparison between characteristics of developed vector algorithms is also presented. © Springer-Verlag Berlin Heidelberg 2005
  6. Keywords:
  7. Algorithms ; Numerical analysis ; Personal computers ; Vectors ; Interpolation algorithms ; Pentium processors ; SIMD ; Vector algorithms ; Interpolation
  8. Source: 1st International Conference on High Performance Computing and Communcations, HPCC 2005, Sorrento, 21 September 2005 through 23 September 2005 ; Volume 3726 LNCS , 2005 , Pages 156-165 ; 03029743 (ISSN); 3540290311 (ISBN); 9783540290315 (ISBN)
  9. URL: https://link.springer.com/chapter/10.1007/11557654_21