Matches in SemOpenAlex for { <https://semopenalex.org/work/W2560804921> ?p ?o ?g. }
Showing items 1 to 68 of
68
with 100 items per page.
- W2560804921 abstract "Upcoming high-performance computing (HPC) platforms will have more complex memory hierarchies with high-bandwidth on-package memory and in the future also non-volatile memory. How to use such deep memory hierarchies effectively remains an open research question. In this paper we evaluate the performance implications of a scheme based on a software-managed scratchpad with coarse-grained memory-copy operations migrating application data structures between memory hierarchy levels. We expect that such a scheme can, under specificcircumstances, outperform a hardware-managed cache while requiring a lot less effort than would a scheme managed entirely by the application programmers. Because suitable hardware is not yet generally available, we propose and benchmark several existing hardware configurations that can be used as approximations, including non-uniform memory access (NUMA) systems and memory on accelerators. We then evaluate data migration mechanisms currently available on Linux systems, such as move_pages and memcpy. We also design a best-case-scenario HPC benchmark to explore how the memory locality and parallelism of applications can be improved by data migration. We find that NUMA systems can be a reasonable approximation platform, especially when auxiliary load mechanisms are employed. Memory migration mechanisms inside the Linux kernel turn out to significantly lag behind a plain user-space memory copy, even after we level the playing field as much as possible. Our dedicated application benchmark demonstrates a significant performance benefit of doing memory migrations—approaching the measured difference in the memory bandwidth—provided that the ratio of worker threads to migration threads is chosen well." @default.
- W2560804921 created "2016-12-16" @default.
- W2560804921 creator A5001838603 @default.
- W2560804921 creator A5021728155 @default.
- W2560804921 creator A5038177319 @default.
- W2560804921 creator A5055862557 @default.
- W2560804921 creator A5090734292 @default.
- W2560804921 date "2016-09-01" @default.
- W2560804921 modified "2023-10-03" @default.
- W2560804921 title "Exploring Data Migration for Future Deep-Memory Many-Core Systems" @default.
- W2560804921 cites W1509960038 @default.
- W2560804921 cites W1548582164 @default.
- W2560804921 cites W1601855692 @default.
- W2560804921 cites W1973948156 @default.
- W2560804921 cites W1983929828 @default.
- W2560804921 cites W1993089791 @default.
- W2560804921 cites W1995064490 @default.
- W2560804921 cites W2036506377 @default.
- W2560804921 cites W2040703170 @default.
- W2560804921 cites W2051446080 @default.
- W2560804921 cites W2054304755 @default.
- W2560804921 cites W2092212481 @default.
- W2560804921 cites W2103881935 @default.
- W2560804921 cites W2105524676 @default.
- W2560804921 cites W2133521520 @default.
- W2560804921 cites W2150662281 @default.
- W2560804921 cites W2153644244 @default.
- W2560804921 cites W2163504658 @default.
- W2560804921 cites W2167098348 @default.
- W2560804921 cites W2171119362 @default.
- W2560804921 cites W2475636809 @default.
- W2560804921 doi "https://doi.org/10.1109/cluster.2016.42" @default.
- W2560804921 hasPublicationYear "2016" @default.
- W2560804921 type Work @default.
- W2560804921 sameAs 2560804921 @default.
- W2560804921 citedByCount "6" @default.
- W2560804921 countsByYear W25608049212017 @default.
- W2560804921 countsByYear W25608049212018 @default.
- W2560804921 countsByYear W25608049212019 @default.
- W2560804921 crossrefType "proceedings-article" @default.
- W2560804921 hasAuthorship W2560804921A5001838603 @default.
- W2560804921 hasAuthorship W2560804921A5021728155 @default.
- W2560804921 hasAuthorship W2560804921A5038177319 @default.
- W2560804921 hasAuthorship W2560804921A5055862557 @default.
- W2560804921 hasAuthorship W2560804921A5090734292 @default.
- W2560804921 hasConcept C2164484 @default.
- W2560804921 hasConcept C41008148 @default.
- W2560804921 hasConcept C76155785 @default.
- W2560804921 hasConceptScore W2560804921C2164484 @default.
- W2560804921 hasConceptScore W2560804921C41008148 @default.
- W2560804921 hasConceptScore W2560804921C76155785 @default.
- W2560804921 hasLocation W25608049211 @default.
- W2560804921 hasOpenAccess W2560804921 @default.
- W2560804921 hasPrimaryLocation W25608049211 @default.
- W2560804921 hasRelatedWork W2093578348 @default.
- W2560804921 hasRelatedWork W2351109918 @default.
- W2560804921 hasRelatedWork W2358668433 @default.
- W2560804921 hasRelatedWork W2376932109 @default.
- W2560804921 hasRelatedWork W2382290278 @default.
- W2560804921 hasRelatedWork W2390279801 @default.
- W2560804921 hasRelatedWork W2748952813 @default.
- W2560804921 hasRelatedWork W2782327182 @default.
- W2560804921 hasRelatedWork W2899084033 @default.
- W2560804921 hasRelatedWork W93493066 @default.
- W2560804921 isParatext "false" @default.
- W2560804921 isRetracted "false" @default.
- W2560804921 magId "2560804921" @default.
- W2560804921 workType "article" @default.