Show simple item record

dc.contributor.advisorBurtscher, Martin
dc.contributor.authorBelcher, Kristi ( Orcid Icon 0000-0002-5774-1796 )
dc.date.accessioned2017-02-09T15:22:04Z
dc.date.available2017-02-09T15:22:04Z
dc.date.issued2016-12
dc.identifier.citationBelcher, K. (2016). Multi-GPU parallelization of irregular algorithms (Unpublished thesis). Texas State University, San Marcos, Texas.
dc.identifier.urihttps://digital.library.txstate.edu/handle/10877/6456
dc.descriptionPresented to the Honors Committee of Texas State University in Partial Fulfillment of the Requirements for Graduation in the University Honors Program, December 2016.
dc.description.abstractAll programs possess a certain degree of irregularity in their control flow and memory ac-cess patterns. The more irregular a program is, the harder it tends to be to parallelize and port to accelerators such as Graphics Processing Units (GPUs). Additionally, efficient ac-celerator-based computing devices are rapidly spreading since they provide more perfor-mance and better energy efficiency than conventional computers. Multi-accelerator sys-tems are already on the horizon and will likely be commonplace in the near future. Hence, it is important to learn how to efficiently run irregular computations on multi-ac-celerator platforms. I have rewritten four single-GPU programs, each with different amounts of irregularity, so that they can exploit multiple GPUs simultaneously. By ana-lyzing shared variables and data dependencies within the programs, I was able to create a general approach for parallelizing programs across multiple accelerators. I then compared the performance of these codes against their single-GPU counterparts to determine the performance benefit and how irregularity impacts that benefit. My results show that mostly regular programs and programs that display control flow irregularity tend to ob-tain a significant performance boost. However, programs that display memory access ir-regularity tend not to gain any speedup from multiple GPUs.en_US
dc.formatText
dc.format.extent23 pages
dc.format.medium1 file (.pdf)
dc.language.isoen
dc.subjectParallel irregularity
dc.subjectMulti-GPU
dc.subjectHigh performance computing
dc.subjectGraphics Processing Uniten_US
dc.subjectMulti-accelerator systemsen_US
dc.subjectControl flow irregularityen_US
dc.subjectMemory access irregularityen_US
dc.titleMulti-GPU Parallelization of Irregular Algorithmsen_US
txstate.documenttypeThesis
thesis.degree.departmentHonors College
thesis.degree.disciplineComputer Science
thesis.degree.grantorTexas State University
txstate.departmentHonors College


Download

Thumbnail

This item appears in the following Collection(s)

Show simple item record