Distributed terascale volume visualization using distributed shared virtual memory

Handle URI:
http://hdl.handle.net/10754/575803
Title:
Distributed terascale volume visualization using distributed shared virtual memory
Authors:
Beyer, Johanna; Hadwiger, Markus ( 0000-0003-1239-4871 ) ; Schneider, Jens; Jeong, Wonki; Pfister, Hanspeter
Abstract:
Table 1 illustrates the impact of different distribution unit sizes, different screen resolutions, and numbers of GPU nodes. We use two and four GPUs (NVIDIA Quadro 5000 with 2.5 GB memory) and a mouse cortex EM dataset (see Figure 2) of resolution 21,494 x 25,790 x 1,850 = 955GB. The size of the virtual distribution units significantly influences the data distribution between nodes. Small distribution units result in a high depth complexity for compositing. Large distribution units lead to a low utilization of GPUs, because in the worst case only a single distribution unit will be in view, which is rendered by only a single node. The choice of an optimal distribution unit size depends on three major factors: the output screen resolution, the block cache size on each node, and the number of nodes. Currently, we are working on optimizing the compositing step and network communication between nodes. © 2011 IEEE.
KAUST Department:
Visual Computing Center (VCC); Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division; Computer Science Program
Publisher:
Institute of Electrical and Electronics Engineers (IEEE)
Journal:
2011 IEEE Symposium on Large Data Analysis and Visualization
Conference/Event name:
1st IEEE Symposium on Large-Scale Data Analysis and Visualization 2011, LDAV 2011
Issue Date:
Oct-2011
DOI:
10.1109/LDAV.2011.6092332
Type:
Conference Paper
ISBN:
9781467301541
Appears in Collections:
Conference Papers; Computer Science Program; Computer Science Program; Visual Computing Center (VCC); Visual Computing Center (VCC); Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorBeyer, Johannaen
dc.contributor.authorHadwiger, Markusen
dc.contributor.authorSchneider, Jensen
dc.contributor.authorJeong, Wonkien
dc.contributor.authorPfister, Hanspeteren
dc.date.accessioned2015-08-24T09:26:37Zen
dc.date.available2015-08-24T09:26:37Zen
dc.date.issued2011-10en
dc.identifier.isbn9781467301541en
dc.identifier.doi10.1109/LDAV.2011.6092332en
dc.identifier.urihttp://hdl.handle.net/10754/575803en
dc.description.abstractTable 1 illustrates the impact of different distribution unit sizes, different screen resolutions, and numbers of GPU nodes. We use two and four GPUs (NVIDIA Quadro 5000 with 2.5 GB memory) and a mouse cortex EM dataset (see Figure 2) of resolution 21,494 x 25,790 x 1,850 = 955GB. The size of the virtual distribution units significantly influences the data distribution between nodes. Small distribution units result in a high depth complexity for compositing. Large distribution units lead to a low utilization of GPUs, because in the worst case only a single distribution unit will be in view, which is rendered by only a single node. The choice of an optimal distribution unit size depends on three major factors: the output screen resolution, the block cache size on each node, and the number of nodes. Currently, we are working on optimizing the compositing step and network communication between nodes. © 2011 IEEE.en
dc.publisherInstitute of Electrical and Electronics Engineers (IEEE)en
dc.titleDistributed terascale volume visualization using distributed shared virtual memoryen
dc.typeConference Paperen
dc.contributor.departmentVisual Computing Center (VCC)en
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.contributor.departmentComputer Science Programen
dc.identifier.journal2011 IEEE Symposium on Large Data Analysis and Visualizationen
dc.conference.date23 October 2011 through 24 October 2011en
dc.conference.name1st IEEE Symposium on Large-Scale Data Analysis and Visualization 2011, LDAV 2011en
dc.conference.locationProvidence, RIen
dc.contributor.institutionHarvard University, United Statesen
kaust.authorBeyer, Johannaen
kaust.authorHadwiger, Markusen
kaust.authorSchneider, Jensen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.