A comparison of multithreading, vectorization, and GPU computing for the acceleration of cardiac electrophysiology models
GUERMOUCHE, Amina
Institut de Mathématiques de Bordeaux [IMB]
STatic Optimizations, Runtime Methods [STORM]
Institut de Mathématiques de Bordeaux [IMB]
STatic Optimizations, Runtime Methods [STORM]
AUMAGE, Olivier
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
STatic Optimizations, Runtime Methods [STORM]
See more >
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
STatic Optimizations, Runtime Methods [STORM]
GUERMOUCHE, Amina
Institut de Mathématiques de Bordeaux [IMB]
STatic Optimizations, Runtime Methods [STORM]
Institut de Mathématiques de Bordeaux [IMB]
STatic Optimizations, Runtime Methods [STORM]
AUMAGE, Olivier
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
STatic Optimizations, Runtime Methods [STORM]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
STatic Optimizations, Runtime Methods [STORM]
SAILLARD, Emmanuelle
STatic Optimizations, Runtime Methods [STORM]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
STatic Optimizations, Runtime Methods [STORM]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
POTSE, Mark
IHU-LIRYC
Modélisation et calculs pour l'électrophysiologie cardiaque [CARMEN]
Institut de Mathématiques de Bordeaux [IMB]
IHU-LIRYC
Modélisation et calculs pour l'électrophysiologie cardiaque [CARMEN]
Institut de Mathématiques de Bordeaux [IMB]
COUDIÈRE, Yves
Modélisation et calculs pour l'électrophysiologie cardiaque [CARMEN]
Institut de Mathématiques de Bordeaux [IMB]
IHU-LIRYC
Modélisation et calculs pour l'électrophysiologie cardiaque [CARMEN]
Institut de Mathématiques de Bordeaux [IMB]
IHU-LIRYC
BARTHOU, Denis
STatic Optimizations, Runtime Methods [STORM]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
< Reduce
STatic Optimizations, Runtime Methods [STORM]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Language
en
Communication dans un congrès
This item was published in
Computing in Cardiology 2022, 2022-09-05, Tampere.
English Abstract
Realistic simulation of cardiac electrophysiology requires both high resolution and computationally expensive models of membrane dynamics. Optimization of membrane models can therefore have a large impact on time, hardware, ...Read more >
Realistic simulation of cardiac electrophysiology requires both high resolution and computationally expensive models of membrane dynamics. Optimization of membrane models can therefore have a large impact on time, hardware, and energy usage. We tested both CPU-based and GPU-based optimization techniques for a human heart model with Ten Tusscher-Panfilov 2006 dynamics. Compared to a multithreaded code running on 64 CPU cores, the tested NVIDIA Tesla P100 GPU proved about 3 times faster. Effective use of the CPU's SIMD capabilities allowed a similar performance gain. GPU performance was bounded by the data transfer rate between GPU and main memory. Optimal SIMD use required explicit vectorization and an adapted data structure. We conclude that on mixed CPU-GPU systems the best results are obtained by optimizing both CPU and GPU code and using a runtime system that balances CPU and GPU load.Read less <
ANR Project
Simulation exascale de l'électrophysiologie cardiaque pour la recherche sur les arythmies - ANR-18-CE46-0010
L'Institut de Rythmologie et modélisation Cardiaque - ANR-10-IAHU-0004
L'Institut de Rythmologie et modélisation Cardiaque - ANR-10-IAHU-0004
Origin
Hal imported