DC Field | Value | Language |
---|---|---|
dc.contributor.author | Jung, Myoungsoo | ko |
dc.contributor.author | Wilson, Ellis H., III | ko |
dc.contributor.author | Choi, Wonil | ko |
dc.contributor.author | Shalf, John | ko |
dc.contributor.author | Aktulga, Hasan Metin | ko |
dc.contributor.author | Yang, Chao | ko |
dc.contributor.author | Saule, Erik | ko |
dc.contributor.author | Catalyurek, Umit V. | ko |
dc.contributor.author | Kandemir, Mahmut | ko |
dc.date.accessioned | 2019-03-19T01:07:11Z | - |
dc.date.available | 2019-03-19T01:07:11Z | - |
dc.date.created | 2019-02-28 | - |
dc.date.issued | 2014 | - |
dc.identifier.citation | SCIENTIFIC PROGRAMMING, v.22, no.2, pp.125 - 139 | - |
dc.identifier.issn | 1058-9244 | - |
dc.identifier.uri | http://hdl.handle.net/10203/251543 | - |
dc.description.abstract | Drawing parallels to the rise of general purpose graphical processing units (GPGPUs) as accelerators for specific high-performance computing (HPC) workloads, there is a rise in the use of non-volatile memory (NVM) as accelerators for I/O-intensive scientific applications. However, existing works have explored use of NVM within dedicated I/O nodes, which are distant from the compute nodes that actually need such acceleration. As NVM bandwidth begins to out-pace point-to-point network capacity, we argue for the need to break from the archetype of completely separated storage. Therefore, in this work we investigate co-location of NVM and compute by varying I/O interfaces, file systems, types of NVM, and both current and future SSD architectures, uncovering numerous bottlenecks implicit in these various levels in the I/O stack. We present novel hardware and software solutions, including the new Unified File System (UFS), to enable fuller utilization of the new compute-local NVM storage. Our experimental evaluation, which employs a real-world Out-of-Core (OoC) HPC application, demonstrates throughput increases in excess of an order of magnitude over current approaches. | - |
dc.language | English | - |
dc.publisher | IOS PRESS | - |
dc.title | Exploring the future of out-of-core computing with compute-local non-volatile memory | - |
dc.type | Article | - |
dc.identifier.wosid | 000336429300006 | - |
dc.identifier.scopusid | 2-s2.0-84901783566 | - |
dc.type.rims | ART | - |
dc.citation.volume | 22 | - |
dc.citation.issue | 2 | - |
dc.citation.beginningpage | 125 | - |
dc.citation.endingpage | 139 | - |
dc.citation.publicationname | SCIENTIFIC PROGRAMMING | - |
dc.identifier.doi | 10.3233/SPR-140384 | - |
dc.contributor.localauthor | Jung, Myoungsoo | - |
dc.contributor.nonIdAuthor | Wilson, Ellis H., III | - |
dc.contributor.nonIdAuthor | Choi, Wonil | - |
dc.contributor.nonIdAuthor | Shalf, John | - |
dc.contributor.nonIdAuthor | Aktulga, Hasan Metin | - |
dc.contributor.nonIdAuthor | Yang, Chao | - |
dc.contributor.nonIdAuthor | Saule, Erik | - |
dc.contributor.nonIdAuthor | Catalyurek, Umit V. | - |
dc.contributor.nonIdAuthor | Kandemir, Mahmut | - |
dc.description.isOpenAccess | N | - |
dc.type.journalArticle | Article | - |
dc.subject.keywordAuthor | High performance computing | - |
dc.subject.keywordAuthor | scientific application | - |
dc.subject.keywordAuthor | non-volatile memory | - |
dc.subject.keywordAuthor | solid state disk | - |
dc.subject.keywordAuthor | out-of-core computing | - |
dc.subject.keywordPlus | FILE SYSTEM | - |
dc.subject.keywordPlus | FLASH | - |
dc.subject.keywordPlus | EIGENSOLVER | - |
dc.subject.keywordPlus | PARALLELISM | - |
dc.subject.keywordPlus | I/O | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.