Accelerating tumour growth simulations on many-core architectures: A case study on the use of GPGPU within VPH

Liu, B; Clapworthy, GJ; Dong, F; Kolokotroni, E; Stamatakos, G

dc.contributor.author	Liu, B	en
dc.contributor.author	Clapworthy, GJ	en
dc.contributor.author	Dong, F	en
dc.contributor.author	Kolokotroni, E	en
dc.contributor.author	Stamatakos, G	en
dc.date.accessioned	2014-03-01T02:47:16Z
dc.date.available	2014-03-01T02:47:16Z
dc.date.issued	2011	en
dc.identifier.issn	10939547	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/33041
dc.subject	CUDA	en
dc.subject	GPGPU	en
dc.subject	in silico oncology	en
dc.subject	multi-GPU	en
dc.subject	Tumour simulation	en
dc.subject	Virtual Physiological Human	en
dc.subject	VPH	en
dc.subject.other	CUDA	en
dc.subject.other	GPGPU	en
dc.subject.other	multi-GPU	en
dc.subject.other	Virtual physiological human	en
dc.subject.other	VPH	en
dc.subject.other	Physiology	en
dc.subject.other	Program processors	en
dc.subject.other	Tumors	en
dc.subject.other	Virtual reality	en
dc.subject.other	Visualization	en
dc.subject.other	Computer graphics equipment	en
dc.title	Accelerating tumour growth simulations on many-core architectures: A case study on the use of GPGPU within VPH	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1109/IV.2011.45	en
heal.identifier.secondary	http://dx.doi.org/10.1109/IV.2011.45	en
heal.identifier.secondary	6004108	en
heal.publicationDate	2011	en
heal.abstract	Simulators of tumour growth can estimate the evolution of tumour volume and the quantity of various categories of cells as functions of time. However, the execution time of each simulation often takes several dozens of minutes (depending upon the dataset resolution), which clearly prevents easy interaction. The modern graphics processing unit (GPU) is not only a powerful graphics engine but also a highly parallel programmable processor featuring peak arithmetic performance and memory bandwidth that substantially outpaces its CPU counterpart. However, despite this, the GPU is little used in the context of the Virtual Physiological Human (VPH). This paper provides a case study to demonstrate the performance advantages that can be gained by using the GPU appropriately in the context of a VPH project in which the study of tumour growth is a central activity. We also analyse the algorithm performance on different modern parallel processing architectures, including multicore CPU and many-core GPU. © 2011 IEEE.	en
heal.journalName	Proceedings of the International Conference on Information Visualisation	en
dc.identifier.doi	10.1109/IV.2011.45	en
dc.identifier.spage	601	en
dc.identifier.epage	609	en