On accelerating iterative algorithms with CUDA: A case study on conditional random fields training algorithm for biological sequence alignment

Zhihui Du, Zhaoming Yin, Wenjie Liu, David Bader

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Scopus citations

Abstract

The accuracy of Conditional Random Fields (CRF) is achieved at the cost of huge amount of computation to train model. In this paper we designed the parallelized algorithm for the Gradient Ascent based CRF training methods for biological sequence alignment. Our contribution is mainly on two aspects: 1) We flexibly parallelized the different iterative computation patterns, and the according optimization methods are presented. 2) As for the Gibbs Sampling based training method, we designed a way to automatically predict the iteration round, so that the parallel algorithm could be run in a more efficient manner. In the experiment, these parallel algorithms achieved valuable accelerations comparing to the serial version.

Original languageEnglish (US)
Title of host publication2010 IEEE International Conference on Bioinformatics and Biomedicine Workshops, BIBMW 2010
Pages543-548
Number of pages6
DOIs
StatePublished - 2010
Externally publishedYes
Event2010 IEEE International Conference on Bioinformatics and Biomedicine Workshops, BIBMW 2010 - HongKong, China
Duration: Dec 18 2010Dec 21 2010

Publication series

Name2010 IEEE International Conference on Bioinformatics and Biomedicine Workshops, BIBMW 2010

Conference

Conference2010 IEEE International Conference on Bioinformatics and Biomedicine Workshops, BIBMW 2010
Country/TerritoryChina
CityHongKong
Period12/18/1012/21/10

All Science Journal Classification (ASJC) codes

  • Biomedical Engineering
  • Health Informatics

Keywords

  • Biological sequence alignment
  • Conditional random fields
  • GPGPU

Fingerprint

Dive into the research topics of 'On accelerating iterative algorithms with CUDA: A case study on conditional random fields training algorithm for biological sequence alignment'. Together they form a unique fingerprint.

Cite this