Results 311 to 320 of about 2,725,619 (370)

Matrix-Matrix Multiplication Using Multiple GPUs Connected by Nvlink

2020 Global Smart Industry Conference (GloSIC), 2020
In this work we present an original GPU-only parallel matrix-matrix multiplication algorithm $(C = aA * B + \beta C)$ for servers with multiple GPUs connected by NVLink. The algorithm is implemented using CUDA. The data transfer patterns, the communication and computation overlap, and the overall performance of the algorithm are considered.
Yea Rem Choi, Vsevolod Nikolskiy, Vladimir Stegailov   +2 more
openaire   +1 more source

More Asymmetry Yields Faster Matrix Multiplication

ACM-SIAM Symposium on Discrete Algorithms
We present a new improvement on the laser method for designing fast matrix multiplication algorithms. The new method further develops the recent advances by [Duan, Wu, Zhou FOCS 2023] and [Vassilevska Williams, Xu, Xu, Zhou SODA 2024].
Josh Alman, Ran Duan, Virginia Vassilevska Williams, Yinzhan Xu, Zixuan Xu, Renfei Zhou   +5 more
semanticscholar   +1 more source

Fault tolerant matrix-matrix multiplication

Proceedings of the second workshop on Scalable algorithms for large-scale systems, 2011
Soft errors are one-time events that corrupt the state of a computing system but not its overall functionality. Soft errors normally do not interrupt the execution of the affected program, but the affected computation results can not be trusted any more.
Panruo Wu, Chong Ding, Longxiang Chen, Feng Gao, Teresa Davies, Christer Karlsson, Zizhong Chen   +6 more
openaire   +1 more source

Multiple BDD based matrix multiplication

2010 IEEE International Conference on Semiconductor Electronics (ICSE2010), 2010
Binary Decision Diagrams (BDDs) are the most frequently used data structure for handling Boolean functions because of their excellent efficiency in terms of time and space. Algebraic Decision Diagrams (ADDs) have been used to solve general purpose problems such as Matrix Multiplication, logic synthesis and Formal Verification. We propose a Multiple BDD
T. Bhuvaneswari, V. C. Prasad, Ajay Kumar Singh   +2 more
openaire   +1 more source

A novel data transformation and execution strategy for accelerating sparse matrix multiplication on GPUs

ACM SIGPLAN Symposium on Principles & Practice of Parallel Programming, 2020
SpMM (multiplication of a sparse matrix and a dense matrix) and SDDMM (sampled dense-dense matrix multiplication) are at the core of many scientific, machine learning, and data mining applications.
Peng Jiang, Changwan Hong, G. Agrawal
semanticscholar   +1 more source

Towards Efficient Sparse Matrix Vector Multiplication on Real Processing-In-Memory Architectures

Measurement and Modeling of Computer Systems, 2022
Several manufacturers have already started to commercialize near-bank Processing-In-Memory (PIM) architectures, after decades of research efforts. Near-bank PIM architectures place simple cores close to DRAM banks.
Christina Giannoula, Ivan Fernandez, Juan Gómez-Luna, N. Koziris, G. Goumas, O. Mutlu   +5 more
semanticscholar   +1 more source

spECK: accelerating GPU sparse matrix-matrix multiplication through lightweight analysis

ACM SIGPLAN Symposium on Principles & Practice of Parallel Programming, 2020
Sparse general matrix-matrix multiplication on GPUs is challenging due to the varying sparsity patterns of sparse matrices. Existing solutions achieve good performance for certain types of matrices, but fail to accelerate all kinds of matrices in the ...
Mathias Parger, Martin Winter, Daniel Mlakar, M. Steinberger   +3 more
semanticscholar   +1 more source

Red-Blue Pebbling Revisited: Near Optimal Parallel Matrix-Matrix Multiplication

International Conference for High Performance Computing, Networking, Storage and Analysis, 2019
We propose COSMA: a parallel matrix-matrix multiplication algorithm that is near communication-optimal for all combinations of matrix dimensions, processor counts, and memory sizes.
Grzegorz Kwasniewski, Marko Kabic, Maciej Besta, J. VandeVondele, R. Solcà, T. Hoefler   +5 more
semanticscholar   +1 more source

Matrix multiplication with DNA

Journal of Molecular Evolution, 1997
A DNA-based method for calculating the product of Boolean matrices or matrices containing positive, real numbers is presented. In the case of matrices containing real numbers, the manipulation of reaction conditions allows a quantitative calculation to be performed. The use of DNA to perform an analog calculation illustrates a new approach to computing
openaire   +2 more sources

Visualizing Matrix Multiplication

PRIMUS, 2017
Efficient visualizations of computational algorithms are important tools for students, educators, and researchers.
Peteris Daugulis, Anita Sondore
openaire   +1 more source