Understanding the Efficiency of GPU Algorithms for Matrix-Matrix Multiplication

Simple item page
dc.contributor.authorFatahalian, K.en_US
dc.contributor.authorSugerman, J.en_US
dc.contributor.authorHanrahan, P.en_US
dc.contributor.editorTomas Akenine-Moeller and Michael McCoolen_US
dc.date.accessioned2013-10-28T10:02:19Z
dc.date.available2013-10-28T10:02:19Z
dc.date.issued2004en_US
dc.description.abstractUtilizing graphics hardware for general purpose numerical computations has become a topic of considerable interest. The implementation of streaming algorithms, typified by highly parallel computations with little reuse of input data, has been widely explored on GPUs. We relax the streaming model's constraint on input reuse and perform an in-depth analysis of dense matrix-matrix multiplication, which reuses each element of input matrices O(n) times. Its regular data access pattern and highly parallel computational requirements suggest matrix-matrix multiplication as an obvious candidate for efficient evaluation on GPUs but, surprisingly we find even nearoptimal GPU implementations are pronouncedly less efficient than current cache-aware CPU approaches. We find the key cause of this inefficiency is that the GPU can fetch less data and yet execute more arithmetic operations per clock than the CPU when both are operating out of their closest caches. The lack of high bandwidth access to cached data will impair the performance of GPU implementations of any computation featuring significant input reuse.en_US
dc.description.seriesinformationGraphics Hardwareen_US
dc.identifier.isbn3-905673-15-0en_US
dc.identifier.issn1727-3471en_US
dc.identifier.urihttps://doi.org/10.2312/EGGH/EGGH04/133-138en_US
dc.publisherThe Eurographics Associationen_US
dc.subjectI.3.1 [Computer Graphics]en_US
dc.subjectGraphics processorsen_US
dc.titleUnderstanding the Efficiency of GPU Algorithms for Matrix-Matrix Multiplicationen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
133-138.pdf
Size:
132.45 KB
Format:
Adobe Portable Document Format
Download
Collections
EGGH04: SIGGRAPH/Eurographics Workshop on Graphics Hardware 2004