Cardiac simulation on multi-GPU platform

Venkata Krishna Nimmagadda, Ali Akoglu, Salim A Hariri, Talal Moukabary

Research output: Contribution to journalArticle

20 Citations (Scopus)

Abstract

The cardiac bidomain model is a popular approach to study electrical behavior of tissues and simulate interactions between the cells by solving partial differential equations. The iterative and data parallel model is an ideal match for the parallel architecture of Graphic Processing Units (GPUs). In this study, we evaluate the effectiveness of architecture-specific optimizations and fine grained parallelization strategies, completely port the model to GPU, and evaluate the performance of single-GPU and multi-GPU implementations. Simulating one action potential duration (350 msec real time) for a 256×256×256 tissue takes 453 hours on a high-end general purpose processor, while it takes 664 seconds on a four-GPU based system including the communication and data transfer overhead. This drastic improvement (a factor of 2460×) will allow clinicians to extend the time-scale of simulations from milliseconds to seconds and minutes; and evaluate hypotheses in a shorter amount of time that was not feasible previously.

Original languageEnglish (US)
Pages (from-to)1360-1378
Number of pages19
JournalJournal of Supercomputing
Volume59
Issue number3
DOIs
StatePublished - 2012

Fingerprint

Graphics Processing Unit
Cardiac
Simulation
Evaluate
Tissue
Parallel architectures
Action Potential
Parallel Architectures
Data Transfer
Data transfer
Parallelization
Partial differential equations
Time Scales
Partial differential equation
Graphics processing unit
Model
Optimization
Communication
Cell
Interaction

Keywords

  • Bidomain model
  • Cardiac tissue
  • Graphics processing unit
  • Multi-GPU

ASJC Scopus subject areas

  • Hardware and Architecture
  • Software
  • Information Systems
  • Theoretical Computer Science

Cite this

Cardiac simulation on multi-GPU platform. / Nimmagadda, Venkata Krishna; Akoglu, Ali; Hariri, Salim A; Moukabary, Talal.

In: Journal of Supercomputing, Vol. 59, No. 3, 2012, p. 1360-1378.

Research output: Contribution to journalArticle

Nimmagadda, Venkata Krishna ; Akoglu, Ali ; Hariri, Salim A ; Moukabary, Talal. / Cardiac simulation on multi-GPU platform. In: Journal of Supercomputing. 2012 ; Vol. 59, No. 3. pp. 1360-1378.
@article{a8f3750f571040afa41145523806304f,
title = "Cardiac simulation on multi-GPU platform",
abstract = "The cardiac bidomain model is a popular approach to study electrical behavior of tissues and simulate interactions between the cells by solving partial differential equations. The iterative and data parallel model is an ideal match for the parallel architecture of Graphic Processing Units (GPUs). In this study, we evaluate the effectiveness of architecture-specific optimizations and fine grained parallelization strategies, completely port the model to GPU, and evaluate the performance of single-GPU and multi-GPU implementations. Simulating one action potential duration (350 msec real time) for a 256×256×256 tissue takes 453 hours on a high-end general purpose processor, while it takes 664 seconds on a four-GPU based system including the communication and data transfer overhead. This drastic improvement (a factor of 2460×) will allow clinicians to extend the time-scale of simulations from milliseconds to seconds and minutes; and evaluate hypotheses in a shorter amount of time that was not feasible previously.",
keywords = "Bidomain model, Cardiac tissue, Graphics processing unit, Multi-GPU",
author = "Nimmagadda, {Venkata Krishna} and Ali Akoglu and Hariri, {Salim A} and Talal Moukabary",
year = "2012",
doi = "10.1007/s11227-010-0540-x",
language = "English (US)",
volume = "59",
pages = "1360--1378",
journal = "Journal of Supercomputing",
issn = "0920-8542",
publisher = "Springer Netherlands",
number = "3",

}

TY - JOUR

T1 - Cardiac simulation on multi-GPU platform

AU - Nimmagadda, Venkata Krishna

AU - Akoglu, Ali

AU - Hariri, Salim A

AU - Moukabary, Talal

PY - 2012

Y1 - 2012

N2 - The cardiac bidomain model is a popular approach to study electrical behavior of tissues and simulate interactions between the cells by solving partial differential equations. The iterative and data parallel model is an ideal match for the parallel architecture of Graphic Processing Units (GPUs). In this study, we evaluate the effectiveness of architecture-specific optimizations and fine grained parallelization strategies, completely port the model to GPU, and evaluate the performance of single-GPU and multi-GPU implementations. Simulating one action potential duration (350 msec real time) for a 256×256×256 tissue takes 453 hours on a high-end general purpose processor, while it takes 664 seconds on a four-GPU based system including the communication and data transfer overhead. This drastic improvement (a factor of 2460×) will allow clinicians to extend the time-scale of simulations from milliseconds to seconds and minutes; and evaluate hypotheses in a shorter amount of time that was not feasible previously.

AB - The cardiac bidomain model is a popular approach to study electrical behavior of tissues and simulate interactions between the cells by solving partial differential equations. The iterative and data parallel model is an ideal match for the parallel architecture of Graphic Processing Units (GPUs). In this study, we evaluate the effectiveness of architecture-specific optimizations and fine grained parallelization strategies, completely port the model to GPU, and evaluate the performance of single-GPU and multi-GPU implementations. Simulating one action potential duration (350 msec real time) for a 256×256×256 tissue takes 453 hours on a high-end general purpose processor, while it takes 664 seconds on a four-GPU based system including the communication and data transfer overhead. This drastic improvement (a factor of 2460×) will allow clinicians to extend the time-scale of simulations from milliseconds to seconds and minutes; and evaluate hypotheses in a shorter amount of time that was not feasible previously.

KW - Bidomain model

KW - Cardiac tissue

KW - Graphics processing unit

KW - Multi-GPU

UR - http://www.scopus.com/inward/record.url?scp=84869777842&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84869777842&partnerID=8YFLogxK

U2 - 10.1007/s11227-010-0540-x

DO - 10.1007/s11227-010-0540-x

M3 - Article

AN - SCOPUS:84869777842

VL - 59

SP - 1360

EP - 1378

JO - Journal of Supercomputing

JF - Journal of Supercomputing

SN - 0920-8542

IS - 3

ER -