Scalable and highly parallel implementation of Smith-Waterman on graphics processing unit using CUDA

Ali Akoglu, Gregory M. Striemer

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

Program development environments have enabled graphics processing units (GPUs) to become an attractive high performance computing platform for the scientific community. A commonly posed problem in computational biology is protein database searching for functional similarities. The most accurate algorithm for sequence alignments is Smith-Waterman (SW). However, due to its computational complexity and rapidly increasing database sizes, the process becomes more and more time consuming making cluster based systems more desirable. Therefore, scalable and highly parallel methods are necessary to make SW a viable solution for life science researchers. In this paper we evaluate how SW fits onto the target GPU architecture by exploring ways to map the program architecture on the processor architecture. We develop new techniques to reduce the memory footprint of the application while exploiting the memory hierarchy of the GPU. With this implementation, GSW, we overcome the on chip memory size constraint, achieving 23× speedup compared to a serial implementation. Results show that as the query length increases our speedup almost stays stable indicating the solid scalability of our approach. Additionally this is a first of a kind implementation which purely runs on the GPU instead of a CPU-GPU integrated environment, making our design suitable for porting onto a cluster of GPUs.

Original languageEnglish (US)
Pages (from-to)341-352
Number of pages12
JournalCluster Computing
Volume12
Issue number3
DOIs
StatePublished - 2009

Fingerprint

Data storage equipment
Program processors
Graphics processing unit
Scalability
Computational complexity
Proteins

Keywords

  • Alignment
  • CUDA
  • Graphics processing unit
  • Parallel
  • Scalable
  • Smith-Waterman

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Software

Cite this

Scalable and highly parallel implementation of Smith-Waterman on graphics processing unit using CUDA. / Akoglu, Ali; Striemer, Gregory M.

In: Cluster Computing, Vol. 12, No. 3, 2009, p. 341-352.

Research output: Contribution to journalArticle

@article{94d26535d1104d339bd60475784acea8,
title = "Scalable and highly parallel implementation of Smith-Waterman on graphics processing unit using CUDA",
abstract = "Program development environments have enabled graphics processing units (GPUs) to become an attractive high performance computing platform for the scientific community. A commonly posed problem in computational biology is protein database searching for functional similarities. The most accurate algorithm for sequence alignments is Smith-Waterman (SW). However, due to its computational complexity and rapidly increasing database sizes, the process becomes more and more time consuming making cluster based systems more desirable. Therefore, scalable and highly parallel methods are necessary to make SW a viable solution for life science researchers. In this paper we evaluate how SW fits onto the target GPU architecture by exploring ways to map the program architecture on the processor architecture. We develop new techniques to reduce the memory footprint of the application while exploiting the memory hierarchy of the GPU. With this implementation, GSW, we overcome the on chip memory size constraint, achieving 23× speedup compared to a serial implementation. Results show that as the query length increases our speedup almost stays stable indicating the solid scalability of our approach. Additionally this is a first of a kind implementation which purely runs on the GPU instead of a CPU-GPU integrated environment, making our design suitable for porting onto a cluster of GPUs.",
keywords = "Alignment, CUDA, Graphics processing unit, Parallel, Scalable, Smith-Waterman",
author = "Ali Akoglu and Striemer, {Gregory M.}",
year = "2009",
doi = "10.1007/s10586-009-0089-8",
language = "English (US)",
volume = "12",
pages = "341--352",
journal = "Cluster Computing",
issn = "1386-7857",
publisher = "Kluwer Academic Publishers",
number = "3",

}

TY - JOUR

T1 - Scalable and highly parallel implementation of Smith-Waterman on graphics processing unit using CUDA

AU - Akoglu, Ali

AU - Striemer, Gregory M.

PY - 2009

Y1 - 2009

N2 - Program development environments have enabled graphics processing units (GPUs) to become an attractive high performance computing platform for the scientific community. A commonly posed problem in computational biology is protein database searching for functional similarities. The most accurate algorithm for sequence alignments is Smith-Waterman (SW). However, due to its computational complexity and rapidly increasing database sizes, the process becomes more and more time consuming making cluster based systems more desirable. Therefore, scalable and highly parallel methods are necessary to make SW a viable solution for life science researchers. In this paper we evaluate how SW fits onto the target GPU architecture by exploring ways to map the program architecture on the processor architecture. We develop new techniques to reduce the memory footprint of the application while exploiting the memory hierarchy of the GPU. With this implementation, GSW, we overcome the on chip memory size constraint, achieving 23× speedup compared to a serial implementation. Results show that as the query length increases our speedup almost stays stable indicating the solid scalability of our approach. Additionally this is a first of a kind implementation which purely runs on the GPU instead of a CPU-GPU integrated environment, making our design suitable for porting onto a cluster of GPUs.

AB - Program development environments have enabled graphics processing units (GPUs) to become an attractive high performance computing platform for the scientific community. A commonly posed problem in computational biology is protein database searching for functional similarities. The most accurate algorithm for sequence alignments is Smith-Waterman (SW). However, due to its computational complexity and rapidly increasing database sizes, the process becomes more and more time consuming making cluster based systems more desirable. Therefore, scalable and highly parallel methods are necessary to make SW a viable solution for life science researchers. In this paper we evaluate how SW fits onto the target GPU architecture by exploring ways to map the program architecture on the processor architecture. We develop new techniques to reduce the memory footprint of the application while exploiting the memory hierarchy of the GPU. With this implementation, GSW, we overcome the on chip memory size constraint, achieving 23× speedup compared to a serial implementation. Results show that as the query length increases our speedup almost stays stable indicating the solid scalability of our approach. Additionally this is a first of a kind implementation which purely runs on the GPU instead of a CPU-GPU integrated environment, making our design suitable for porting onto a cluster of GPUs.

KW - Alignment

KW - CUDA

KW - Graphics processing unit

KW - Parallel

KW - Scalable

KW - Smith-Waterman

UR - http://www.scopus.com/inward/record.url?scp=70349595099&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70349595099&partnerID=8YFLogxK

U2 - 10.1007/s10586-009-0089-8

DO - 10.1007/s10586-009-0089-8

M3 - Article

VL - 12

SP - 341

EP - 352

JO - Cluster Computing

JF - Cluster Computing

SN - 1386-7857

IS - 3

ER -