Show simple item record

dc.contributor.advisorVan de Geijn, Robert A.
dc.creatorHuang, Jianyu
dc.date.accessioned2018-10-18T15:14:07Z
dc.date.available2018-10-18T15:14:07Z
dc.date.created2018-08
dc.date.issued2018-10-08
dc.date.submittedAugust 2018
dc.identifierdoi:10.15781/T2V11W511
dc.identifier.urihttp://hdl.handle.net/2152/69013
dc.description.abstractMatrix multiplication is a core building block for numerous scientific computing and, more recently, machine learning applications. Strassen's algorithm, the original Fast Matrix Multiplication (FMM) algorithm, has long fascinated computer scientists due to its startling property of reducing the number of computations required for multiplying n x n matrices from O(n³) to O(n [superscript 2.807]). Over the last half century, this has fueled many theoretical improvements such as other variations of Strassen-like FMM algorithms. Previous implementations of these FMM algorithms led to the "street wisdom" that they are only practical for large, relatively square matrices, that they require considerable workspace, and that they are difficult to achieve thread-level parallelism. The thesis of this work dispels these notions by demonstrating significant benefits for small and non-square matrices, requiring no workspace beyond what is already incorporated in high-performance implementations of matrix multiplication, and achieving performance benefits on multi-core, many-core, and distributed memory architectures.
dc.format.mimetypeapplication/pdf
dc.language.isoen
dc.subjectHigh performance computing
dc.subjectMatrix multiplication
dc.subjectStrassen's algorithm
dc.subjectNumerical algorithm
dc.subjectMathematical software
dc.subjectLinear algebra library
dc.subjectBLAS
dc.subjectPerformance model
dc.subjectTensor contraction
dc.subjectGPU
dc.subjectPerformance optimization
dc.subjectMachine learning
dc.subjectParallel computing
dc.titlePractical fast matrix multiplication algorithms
dc.typeThesis
dc.date.updated2018-10-18T15:14:07Z
dc.contributor.committeeMemberBatory, Don S
dc.contributor.committeeMemberBiros, George
dc.contributor.committeeMemberHenry, Greg M
dc.contributor.committeeMemberPingali, Keshav K
dc.description.departmentComputer Sciences
thesis.degree.departmentComputer Sciences
thesis.degree.disciplineComputer Science
thesis.degree.grantorThe University of Texas at Austin
thesis.degree.levelDoctoral
thesis.degree.nameDoctor of Philosophy
dc.creator.orcid0000-0001-7595-5539
dc.type.materialtext


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record