Large scale cardiac modeling on the Blue Gene supercomputer

Matthias Reumann, Blake G. Fitch, Aleksandr Rayshubskiy, David U. Keller, Daniel L. Weiss, Gunnar Seemann, Olaf Dössel, Michael C. Pitman, John J. Rice

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

Abstract

Multi-scale, multi-physical heart models have not yet been able to include a high degree of accuracy and resolution with respect to model detail and spatial resolution due to computational limitations of current systems. We propose a framework to compute large scale cardiac models. Decomposition of anatomical data in segments to be distributed on a parallel computer is carried out by optimal recursive bisection (ORB). The algorithm takes into account a computational load parameter which has to be adjusted according to the cell models used. The diffusion term is realized by the monodomain equations. The anatomical data-set was given by both ventricles of the Visible Female data-set in a 0.2 mm resolution. Heterogeneous anisotropy was included in the computation. Model weights as input for the decomposition and load balancing were set to (a) 1 for tissue and 0 for non-tissue elements; (b) 10 for tissue and 1 for non-tissue elements. Scaling results for 512, 1024, 2048, 4096 and 8192 computational nodes were obtained for 10 ms simulation time. The simulations were carried out on an IBM Blue Gene/L parallel computer. A 1 s simulation was then carried out on 2048 nodes for the optimal model load. Load balances did not differ significantly across computational nodes even if the number of data elements distributed to each node differed greatly. Since the ORB algorithm did not take into account computational load due to communication cycles, the speedup is close to optimal for the computation time but not optimal overall due to the communication overhead. However, the simulation times were reduced form 87 minutes on 512 to 11 minutes on 8192 nodes. This work demonstrates that it is possible to run simulations of the presented detailed cardiac model within hours for the simulation of a heart beat.

Original languageEnglish (US)
Title of host publicationProceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS'08 - "Personalized Healthcare through Technology"
Pages577-580
Number of pages4
StatePublished - 2008
Externally publishedYes
Event30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS'08 - Vancouver, BC, Canada
Duration: Aug 20 2008Aug 25 2008

Other

Other30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS'08
CountryCanada
CityVancouver, BC
Period8/20/088/25/08

Fingerprint

Supercomputers
Genes
Anisotropy
Weights and Measures
Tissue
Decomposition
Communication
Resource allocation
Datasets

Keywords

  • Computational biology
  • Multi-physical heart models
  • Optimal recursive bisection
  • Parallel supercomputer

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Biomedical Engineering
  • Health Informatics

Cite this

Reumann, M., Fitch, B. G., Rayshubskiy, A., Keller, D. U., Weiss, D. L., Seemann, G., ... Rice, J. J. (2008). Large scale cardiac modeling on the Blue Gene supercomputer. In Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS'08 - "Personalized Healthcare through Technology" (pp. 577-580). [4649218]

Large scale cardiac modeling on the Blue Gene supercomputer. / Reumann, Matthias; Fitch, Blake G.; Rayshubskiy, Aleksandr; Keller, David U.; Weiss, Daniel L.; Seemann, Gunnar; Dössel, Olaf; Pitman, Michael C.; Rice, John J.

Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS'08 - "Personalized Healthcare through Technology". 2008. p. 577-580 4649218.

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

Reumann, M, Fitch, BG, Rayshubskiy, A, Keller, DU, Weiss, DL, Seemann, G, Dössel, O, Pitman, MC & Rice, JJ 2008, Large scale cardiac modeling on the Blue Gene supercomputer. in Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS'08 - "Personalized Healthcare through Technology"., 4649218, pp. 577-580, 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS'08, Vancouver, BC, Canada, 8/20/08.
Reumann M, Fitch BG, Rayshubskiy A, Keller DU, Weiss DL, Seemann G et al. Large scale cardiac modeling on the Blue Gene supercomputer. In Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS'08 - "Personalized Healthcare through Technology". 2008. p. 577-580. 4649218
Reumann, Matthias ; Fitch, Blake G. ; Rayshubskiy, Aleksandr ; Keller, David U. ; Weiss, Daniel L. ; Seemann, Gunnar ; Dössel, Olaf ; Pitman, Michael C. ; Rice, John J. / Large scale cardiac modeling on the Blue Gene supercomputer. Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS'08 - "Personalized Healthcare through Technology". 2008. pp. 577-580
@inproceedings{316602fee4b74b8e8da31480632761d9,
title = "Large scale cardiac modeling on the Blue Gene supercomputer",
abstract = "Multi-scale, multi-physical heart models have not yet been able to include a high degree of accuracy and resolution with respect to model detail and spatial resolution due to computational limitations of current systems. We propose a framework to compute large scale cardiac models. Decomposition of anatomical data in segments to be distributed on a parallel computer is carried out by optimal recursive bisection (ORB). The algorithm takes into account a computational load parameter which has to be adjusted according to the cell models used. The diffusion term is realized by the monodomain equations. The anatomical data-set was given by both ventricles of the Visible Female data-set in a 0.2 mm resolution. Heterogeneous anisotropy was included in the computation. Model weights as input for the decomposition and load balancing were set to (a) 1 for tissue and 0 for non-tissue elements; (b) 10 for tissue and 1 for non-tissue elements. Scaling results for 512, 1024, 2048, 4096 and 8192 computational nodes were obtained for 10 ms simulation time. The simulations were carried out on an IBM Blue Gene/L parallel computer. A 1 s simulation was then carried out on 2048 nodes for the optimal model load. Load balances did not differ significantly across computational nodes even if the number of data elements distributed to each node differed greatly. Since the ORB algorithm did not take into account computational load due to communication cycles, the speedup is close to optimal for the computation time but not optimal overall due to the communication overhead. However, the simulation times were reduced form 87 minutes on 512 to 11 minutes on 8192 nodes. This work demonstrates that it is possible to run simulations of the presented detailed cardiac model within hours for the simulation of a heart beat.",
keywords = "Computational biology, Multi-physical heart models, Optimal recursive bisection, Parallel supercomputer",
author = "Matthias Reumann and Fitch, {Blake G.} and Aleksandr Rayshubskiy and Keller, {David U.} and Weiss, {Daniel L.} and Gunnar Seemann and Olaf D{\"o}ssel and Pitman, {Michael C.} and Rice, {John J.}",
year = "2008",
language = "English (US)",
isbn = "9781424418152",
pages = "577--580",
booktitle = "Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS'08 - {"}Personalized Healthcare through Technology{"}",

}

TY - GEN

T1 - Large scale cardiac modeling on the Blue Gene supercomputer

AU - Reumann, Matthias

AU - Fitch, Blake G.

AU - Rayshubskiy, Aleksandr

AU - Keller, David U.

AU - Weiss, Daniel L.

AU - Seemann, Gunnar

AU - Dössel, Olaf

AU - Pitman, Michael C.

AU - Rice, John J.

PY - 2008

Y1 - 2008

N2 - Multi-scale, multi-physical heart models have not yet been able to include a high degree of accuracy and resolution with respect to model detail and spatial resolution due to computational limitations of current systems. We propose a framework to compute large scale cardiac models. Decomposition of anatomical data in segments to be distributed on a parallel computer is carried out by optimal recursive bisection (ORB). The algorithm takes into account a computational load parameter which has to be adjusted according to the cell models used. The diffusion term is realized by the monodomain equations. The anatomical data-set was given by both ventricles of the Visible Female data-set in a 0.2 mm resolution. Heterogeneous anisotropy was included in the computation. Model weights as input for the decomposition and load balancing were set to (a) 1 for tissue and 0 for non-tissue elements; (b) 10 for tissue and 1 for non-tissue elements. Scaling results for 512, 1024, 2048, 4096 and 8192 computational nodes were obtained for 10 ms simulation time. The simulations were carried out on an IBM Blue Gene/L parallel computer. A 1 s simulation was then carried out on 2048 nodes for the optimal model load. Load balances did not differ significantly across computational nodes even if the number of data elements distributed to each node differed greatly. Since the ORB algorithm did not take into account computational load due to communication cycles, the speedup is close to optimal for the computation time but not optimal overall due to the communication overhead. However, the simulation times were reduced form 87 minutes on 512 to 11 minutes on 8192 nodes. This work demonstrates that it is possible to run simulations of the presented detailed cardiac model within hours for the simulation of a heart beat.

AB - Multi-scale, multi-physical heart models have not yet been able to include a high degree of accuracy and resolution with respect to model detail and spatial resolution due to computational limitations of current systems. We propose a framework to compute large scale cardiac models. Decomposition of anatomical data in segments to be distributed on a parallel computer is carried out by optimal recursive bisection (ORB). The algorithm takes into account a computational load parameter which has to be adjusted according to the cell models used. The diffusion term is realized by the monodomain equations. The anatomical data-set was given by both ventricles of the Visible Female data-set in a 0.2 mm resolution. Heterogeneous anisotropy was included in the computation. Model weights as input for the decomposition and load balancing were set to (a) 1 for tissue and 0 for non-tissue elements; (b) 10 for tissue and 1 for non-tissue elements. Scaling results for 512, 1024, 2048, 4096 and 8192 computational nodes were obtained for 10 ms simulation time. The simulations were carried out on an IBM Blue Gene/L parallel computer. A 1 s simulation was then carried out on 2048 nodes for the optimal model load. Load balances did not differ significantly across computational nodes even if the number of data elements distributed to each node differed greatly. Since the ORB algorithm did not take into account computational load due to communication cycles, the speedup is close to optimal for the computation time but not optimal overall due to the communication overhead. However, the simulation times were reduced form 87 minutes on 512 to 11 minutes on 8192 nodes. This work demonstrates that it is possible to run simulations of the presented detailed cardiac model within hours for the simulation of a heart beat.

KW - Computational biology

KW - Multi-physical heart models

KW - Optimal recursive bisection

KW - Parallel supercomputer

UR - http://www.scopus.com/inward/record.url?scp=61849112666&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=61849112666&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:61849112666

SN - 9781424418152

SP - 577

EP - 580

BT - Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS'08 - "Personalized Healthcare through Technology"

ER -