Results 1 to 10 of about 3,401,789 (270)
Adaptive sparse tiling for sparse matrix multiplication [PDF]
Proceedings of the 24th Symposium on Principles and Practice of Parallel Programming, 2019 Tiling is a key technique for data locality optimization and is widely used in high-performance implementations of dense matrix-matrix multiplication for multicore/manycore CPUs and GPUs. However, the irregular and matrix-dependent data access pattern of sparse matrix multiplication makes it challenging to use tiling to enhance data reuse.
Changwan Hong +4 more
semanticscholar +3 more sources
Sympiler: Transforming Sparse Matrix Codes by Decoupling Symbolic Analysis [PDF]
International Conference for High Performance Computing, Networking, Storage and Analysis, 2017 Sympiler is a domain-specific code generator that optimizes sparse matrix computations by decoupling the symbolic analysis phase from the numerical manipulation stage in sparse codes. The computation patterns in sparse numerical methods are guided by the
Cheshmi, Kazem +3 more
core +2 more sources
Sparse Matrix Factorization [PDF]
, 2014 We investigate the problem of factorizing a matrix into several sparse matrices and propose an algorithm for this under randomness and sparsity assumptions.
Neyshabur, Behnam, Panigrahy, Rina
core +2 more sources
Multi
Concurrency and Computation: Practice and ExperienceABSTRACT The paper focuses on the improvement of the existing nsparse Nagasaka et al. algorithm and its extension to the multi‐GPU setting for the application of real engineering problems.
Artem Mavliutov +4 more
openaire +2 more sources
Efficient Quantized Sparse Matrix Operations on Tensor Cores [PDF]
International Conference for High Performance Computing, Networking, Storage and Analysis, 2022 The exponentially growing model size drives the continued success of deep learning, but it brings prohibitive computation and memory cost. From the algorithm perspective, model sparsification and quantization have been studied to alleviate the problem ...
Shigang Li, Kazuki Osawa, T. Hoefler
semanticscholar +1 more source
Sextans: A Streaming Accelerator for General-Purpose Sparse-Matrix Dense-Matrix Multiplication [PDF]
Symposium on Field Programmable Gate Arrays, 2021 Sparse-Matrix Dense-Matrix multiplication (SpMM) is the key operator for a wide range of applications including scientific computing, graph processing, and deep learning.
Linghao Song +5 more
semanticscholar +1 more source
Spada: Accelerating Sparse Matrix Multiplication with Adaptive Dataflow
International Conference on Architectural Support for Programming Languages and Operating Systems, 2023 Sparse matrix-matrix multiplication (SpGEMM) is widely used in many scientific and deep learning applications. The highly irregular structures of SpGEMM limit its performance and efficiency on conventional computation platforms, and thus motivate a large
Zhiyao Li +6 more
semanticscholar +1 more source
SpArch: Efficient Architecture for Sparse Matrix Multiplication [PDF]
International Symposium on High-Performance Computer Architecture, 2020 Generalized Sparse Matrix-Matrix Multiplication (SpGEMM) is a ubiquitous task in various engineering and scientific applications. However, inner product based SpGEMM introduces redundant input fetches for mismatched nonzero operands, while outer product ...
Zhekai Zhang +3 more
semanticscholar +1 more source
Gamma: leveraging Gustavson’s algorithm to accelerate sparse matrix multiplication
International Conference on Architectural Support for Programming Languages and Operating Systems, 2021 Sparse matrix-sparse matrix multiplication (spMspM) is at the heart of a wide range of scientific and machine learning applications. spMspM is inefficient on general-purpose architectures, making accelerators attractive.
Guowei Zhang +3 more
semanticscholar +1 more source
GE-SpMM: General-Purpose Sparse Matrix-Matrix Multiplication on GPUs for Graph Neural Networks [PDF]
International Conference for High Performance Computing, Networking, Storage and Analysis, 2020 The acceleration of Graph Neural Networks (GNNs) requires efficient and framework-compatible Sparse-Dense Matrix-Matrix Multiplication (SpMM). From the compatibility perspective, the sophisticated sparse matrix representations in state-of-the-art SpMM ...
Guyue Huang +3 more
semanticscholar +1 more source