Zobrazeno 1 - 5
of 5
pro vyhledávání: '"Bob Cernohous"'
Publikováno v:
ACM Transactions on Parallel Computing. 7:1-25
Languages and libraries based on Partitioned Global Address Space (PGAS) programming models are convenient for exploiting scalable parallelism on large applications across different domains with irregular memory access patterns. OpenSHMEM is a PGAS-b
Publikováno v:
OpenSHMEM and Related Technologies. OpenSHMEM in the Era of Extreme Heterogeneity ISBN: 9783030049171
OpenSHMEM
OpenSHMEM
SHMEM has a long history as a parallel programming model. It is extensively used since 1993, starting from Cray T3D systems. For the past two decades SHMEM library implementation in Cray systems evolved through different generations. The current gene
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::7446043c1a0fd0209e8d98b44b37b485
https://doi.org/10.1007/978-3-030-04918-8_3
https://doi.org/10.1007/978-3-030-04918-8_3
Autor:
Naveen Namashivayam, James Dinan, Krishna Kandalla, Mark Pagel, Joseph Robichaux, Dan Pou, Bob Cernohous
Publikováno v:
Lecture Notes in Computer Science ISBN: 9783319738130
OpenSHMEM
OpenSHMEM
To extract best performance from emerging tiered memory systems, it is essential for applications to use the different kinds of memory available on the system. OpenSHMEM memory model consists of data objects that are private to each Processing Elemen
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::6f0eb67791201ac644fc3fdad4984ef9
https://doi.org/10.1007/978-3-319-73814-7_1
https://doi.org/10.1007/978-3-319-73814-7_1
Publikováno v:
OpenSHMEM and Related Technologies. Enhancing OpenSHMEM for Hybrid Environments ISBN: 9783319509945
OpenSHMEM
OpenSHMEM
There is increasing use of multithreading in High Performance Computing (HPC) programs in order to maximize the use of hardware resources of multi-core compute nodes. The latest version of the OpenSHMEM API Specification does not standardize the inte
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::bbb700d0172b088abef3b83ee70b7382
https://doi.org/10.1007/978-3-319-50995-2_11
https://doi.org/10.1007/978-3-319-50995-2_11
Publikováno v:
Concurrency and Computation: Practice and Experience. 30:e4332
Summary Recent Cray-authored Lustre modifications known as Lustre Lockahead show significantly improved write performance for collective, shared-file I/O workloads. Initial tests show write performance improvements of more than 200% for small transfe