Zobrazeno 1 - 10
of 106
pro vyhledávání: '"Sparse matrix computations"'
Publikováno v:
Lecture Notes in Computer Science ISBN: 9783030794774
NPC
NPC
Segmented operations, such as segmented sum, segmented scan and segmented sort, are important building blocks for parallel irregular algorithms. We in this work propose a new parallel primitive called segmented merge. Its function is in parallel merg
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::c5b27de1b182bc574dedd6493b193716
https://doi.org/10.1007/978-3-030-79478-1_15
https://doi.org/10.1007/978-3-030-79478-1_15
Publikováno v:
SC
SC 2020-International Conference for High Performance Computing, Networking, Storage and Analysis
SC 2020-International Conference for High Performance Computing, Networking, Storage and Analysis, Nov 2020, virtual, United States. pp.1-13
SC 2020-International Conference for High Performance Computing, Networking, Storage and Analysis
SC 2020-International Conference for High Performance Computing, Networking, Storage and Analysis, Nov 2020, virtual, United States. pp.1-13
International audience; Tiling is a key technique to reduce data movement in matrix computations. While tiling is well understood and widely used for dense matrix/tensor computations, effective tiling of sparse matrix computations remains a challengi
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c6f98727951467f34222876139f9c413
https://doi.org/10.1109/sc41405.2020.00091
https://doi.org/10.1109/sc41405.2020.00091
Publikováno v:
PACT
Weight pruning is a popular technique to reduce the size and computation complexity of the Convolutional Neural Networks (CNNs). Despite its success in reducing the model size, weight pruning has brought limited benefit to the CNN inference performan
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::fa63012b8fb08e5fd410397c5be94963
https://doi.org/10.1145/3410463.3414648
https://doi.org/10.1145/3410463.3414648
Autor:
Francois-Henry Rouet, Cleve Ashcraft, James Ong, Erman Guleryuz, Todd A. Simons, Ting-Ting Zhu, Seid Koric, Robert F. Lucas, Jef Dawson, Roger G. Grimes
Publikováno v:
IPDPS
LS-DYNA is a well-known multiphysics code with both explicit and implicit time stepping capabilities. Implicit simulations rely heavily on sparse matrix computations, in particular direct solvers, and are notoriously much harder to scale than explici
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::2275322336c640271024fe4f838b2a8b
https://doi.org/10.1109/ipdps47924.2020.00059
https://doi.org/10.1109/ipdps47924.2020.00059
Autor:
Piyush Sao, Ramakrishnan Kannan
Publikováno v:
Parallel Processing and Applied Mathematics ISBN: 9783030432287
PPAM (1)
PPAM (1)
Non-negative matrix factorization (Nmf) is an important tool in high-performance large scale data analytics with applications ranging from community detection, recommender system, feature detection and linear and non-linear unmixing. While traditiona
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::0f7ba548efd34c000dd8120d173c5c18
https://doi.org/10.1007/978-3-030-43229-4_46
https://doi.org/10.1007/978-3-030-43229-4_46
Publikováno v:
HiPC
Data movement is an important bottleneck against efficiency and energy consumption in large-scale sparse matrix computations that are commonly used in linear solvers, eigensolvers and graph analytics. We introduce a novel task-parallel sparse solver
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::a4b49e27147958c06fc34cd311fced47
https://doi.org/10.1109/hipc.2019.00052
https://doi.org/10.1109/hipc.2019.00052
Autor:
Gauri Joshi, Ankur Mallick
Publikováno v:
ISIT
Unpredictable slowdown of worker nodes, or node straggling, is a major bottleneck when performing large matrix computations such as matrix-vector multiplication in a distributed fashion. For sparse matrices, the problem is compounded by irregularitie
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::5d7c12aeaa8dea9a83b744d6a8780632
https://doi.org/10.1109/isit.2019.8849306
https://doi.org/10.1109/isit.2019.8849306
Publikováno v:
SC
In this work, we describe ParSy, a framework that uses a novel inspection strategy along with a simple code transformation to optimize parallel sparse algorithms for shared memory processors. Unlike existing approaches that can suffer from load imbal
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::f67c54971444b16e7555316acebb3717
https://doi.org/10.1109/sc.2018.00065
https://doi.org/10.1109/sc.2018.00065
Publikováno v:
FedCSIS
Due to the ever changing characteristics of the newly provided hardware, there is the permanent requirement of designing and re-designing software adequately to meet the basic hardware conditions. Especially for well-established software, easy portab
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::0ce43e96ab34664b910869abccaa4394
https://doi.org/10.15439/2018f169
https://doi.org/10.15439/2018f169
Publikováno v:
SBAC-PAD
Time-stepping simulation methods offer potential for self-adaptivity, since the first time steps of the simulation can be used to explore the hardware characteristics and measure which of several available implementation variants leads to a good perf
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::c9d16928671d2603cada2a3291a01026
https://doi.org/10.1109/cahpc.2018.8645887
https://doi.org/10.1109/cahpc.2018.8645887