Matches in SemOpenAlex for { <https://semopenalex.org/work/W4366259332> ?p ?o ?g. }
Showing items 1 to 94 of
94
with 100 items per page.
- W4366259332 abstract "Generalized matrix multiplication (GEMM) involves computing the matrix product of any two matrices with appropriate dimensions. Specifically, GEMM doesn't enforce rules on the structure of the entries of the input matrices, such as requiring them to be diagonal, symmetric, or any other special case. Similarly, GEMM doesn't require the matrices to be of a certain density or sparsity. GEMM is utilized in many practical applications including deep learning with convolutional networks, computer vision, and large-scale signal processing. The advantage of a generalized matrix multiplication algorithm is the ability to process big matrix datasets through efficient memory access techniques with a lower requirement for temporary storage than other methods. We designed a parallel divide and conquer general matrix multiplication (PDCGMM) algorithm that performs GEMM comparably for both sparse and dense matrices. PDCGMM also takes advantage of the parallel processing ability of GPUs by using the Computer Unified Device Architecture (CUDA) and efficient usage of GPU memory. We experimented with PDCGMM on five matrices with different sizes and densities to evaluate the algorithm's performance. PDCGMM demonstrated 5–6 times speedup over NumPy's built-in GEMM algorithm for large matrices and 70–90 times speedup for matrices that could be stored entirely on GPU memory." @default.
- W4366259332 created "2023-04-20" @default.
- W4366259332 creator A5013728548 @default.
- W4366259332 creator A5038325765 @default.
- W4366259332 creator A5038553198 @default.
- W4366259332 creator A5043195707 @default.
- W4366259332 creator A5068730526 @default.
- W4366259332 creator A5023860633 @default.
- W4366259332 date "2023-03-08" @default.
- W4366259332 modified "2023-09-27" @default.
- W4366259332 title "An Efficient Parallel Divide-and-Conquer Algorithm for Generalized Matrix Multiplication" @default.
- W4366259332 cites W2035080386 @default.
- W4366259332 cites W2558367073 @default.
- W4366259332 cites W2810319932 @default.
- W4366259332 cites W2889000703 @default.
- W4366259332 cites W2914631005 @default.
- W4366259332 cites W2984305089 @default.
- W4366259332 cites W2991137062 @default.
- W4366259332 cites W3009870468 @default.
- W4366259332 cites W3120865514 @default.
- W4366259332 cites W3132517792 @default.
- W4366259332 cites W3162143314 @default.
- W4366259332 cites W3202268066 @default.
- W4366259332 doi "https://doi.org/10.1109/ccwc57344.2023.10099141" @default.
- W4366259332 hasPublicationYear "2023" @default.
- W4366259332 type Work @default.
- W4366259332 citedByCount "0" @default.
- W4366259332 crossrefType "proceedings-article" @default.
- W4366259332 hasAuthorship W4366259332A5013728548 @default.
- W4366259332 hasAuthorship W4366259332A5023860633 @default.
- W4366259332 hasAuthorship W4366259332A5038325765 @default.
- W4366259332 hasAuthorship W4366259332A5038553198 @default.
- W4366259332 hasAuthorship W4366259332A5043195707 @default.
- W4366259332 hasAuthorship W4366259332A5068730526 @default.
- W4366259332 hasConcept C106487976 @default.
- W4366259332 hasConcept C113313756 @default.
- W4366259332 hasConcept C11413529 @default.
- W4366259332 hasConcept C114614502 @default.
- W4366259332 hasConcept C120373497 @default.
- W4366259332 hasConcept C121332964 @default.
- W4366259332 hasConcept C130367717 @default.
- W4366259332 hasConcept C159985019 @default.
- W4366259332 hasConcept C163716315 @default.
- W4366259332 hasConcept C17349429 @default.
- W4366259332 hasConcept C173608175 @default.
- W4366259332 hasConcept C192562407 @default.
- W4366259332 hasConcept C2524010 @default.
- W4366259332 hasConcept C2778119891 @default.
- W4366259332 hasConcept C2780595030 @default.
- W4366259332 hasConcept C33923547 @default.
- W4366259332 hasConcept C41008148 @default.
- W4366259332 hasConcept C56372850 @default.
- W4366259332 hasConcept C62520636 @default.
- W4366259332 hasConcept C68339613 @default.
- W4366259332 hasConcept C71559656 @default.
- W4366259332 hasConcept C84114770 @default.
- W4366259332 hasConceptScore W4366259332C106487976 @default.
- W4366259332 hasConceptScore W4366259332C113313756 @default.
- W4366259332 hasConceptScore W4366259332C11413529 @default.
- W4366259332 hasConceptScore W4366259332C114614502 @default.
- W4366259332 hasConceptScore W4366259332C120373497 @default.
- W4366259332 hasConceptScore W4366259332C121332964 @default.
- W4366259332 hasConceptScore W4366259332C130367717 @default.
- W4366259332 hasConceptScore W4366259332C159985019 @default.
- W4366259332 hasConceptScore W4366259332C163716315 @default.
- W4366259332 hasConceptScore W4366259332C17349429 @default.
- W4366259332 hasConceptScore W4366259332C173608175 @default.
- W4366259332 hasConceptScore W4366259332C192562407 @default.
- W4366259332 hasConceptScore W4366259332C2524010 @default.
- W4366259332 hasConceptScore W4366259332C2778119891 @default.
- W4366259332 hasConceptScore W4366259332C2780595030 @default.
- W4366259332 hasConceptScore W4366259332C33923547 @default.
- W4366259332 hasConceptScore W4366259332C41008148 @default.
- W4366259332 hasConceptScore W4366259332C56372850 @default.
- W4366259332 hasConceptScore W4366259332C62520636 @default.
- W4366259332 hasConceptScore W4366259332C68339613 @default.
- W4366259332 hasConceptScore W4366259332C71559656 @default.
- W4366259332 hasConceptScore W4366259332C84114770 @default.
- W4366259332 hasLocation W43662593321 @default.
- W4366259332 hasOpenAccess W4366259332 @default.
- W4366259332 hasPrimaryLocation W43662593321 @default.
- W4366259332 hasRelatedWork W1528961374 @default.
- W4366259332 hasRelatedWork W1963713629 @default.
- W4366259332 hasRelatedWork W2030992542 @default.
- W4366259332 hasRelatedWork W2031460602 @default.
- W4366259332 hasRelatedWork W2034258707 @default.
- W4366259332 hasRelatedWork W2256144436 @default.
- W4366259332 hasRelatedWork W2380707708 @default.
- W4366259332 hasRelatedWork W3129397352 @default.
- W4366259332 hasRelatedWork W79781788 @default.
- W4366259332 hasRelatedWork W2130342641 @default.
- W4366259332 isParatext "false" @default.
- W4366259332 isRetracted "false" @default.
- W4366259332 workType "article" @default.