Identifying Opportunities for Byte-Addressable Non-Volatile Memory in Extreme-Scale Scientific Applications

Future exascale systems face extreme power challenges. To improve power efficiency of future HPC systems, non-volatile memory (NVRAM) technologies are being investigated as potential alternatives to existing memories technologies. NVRAMs use extremely low power when in standby mode, and have other p...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2012 IEEE 26th International Parallel and Distributed Processing Symposium s. 945 - 956
Hlavní autoři: Dong Li, Vetter, J. S., Marin, G., McCurdy, C., Cira, C., Zhuo Liu, Weikuan Yu
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.05.2012
Témata:
ISBN:1467309753, 9781467309752
ISSN:1530-2075
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract Future exascale systems face extreme power challenges. To improve power efficiency of future HPC systems, non-volatile memory (NVRAM) technologies are being investigated as potential alternatives to existing memories technologies. NVRAMs use extremely low power when in standby mode, and have other performance and scaling benefits. Although previous work has explored the integration of NVRAM into various architecture and system levels, an open question remains: do specific memory workload characteristics of scientific applications map well onto NVRAMs' features when used in a hybrid NVRAM-DRAM memory system? Furthermore, are there common classes of data structures used by scientific applications that should be frequently placed into NVRAM?In this paper, we analyze several mission-critical scientific applications in order to answer these questions. Specifically, we develop a binary instrumentation tool to statistically report memory access patterns in stack, heap, and global data. We carry out hardware simulation to study the impact of NVRAM for both memory power and system performance. Our study identifies many opportunities for using NVRAM for scientific applications. In two of our applications, 31% and 27% of the memory working sets are suitable for NVRAM. Our simulations suggest at least 27% possible power savings and reveal that the performance of some applications is insensitive to relatively long NVRAM write-access latencies.
AbstractList Future exascale systems face extreme power challenges. To improve power efficiency of future HPC systems, non-volatile memory (NVRAM) technologies are being investigated as potential alternatives to existing memories technologies. NVRAMs use extremely low power when in standby mode, and have other performance and scaling benefits. Although previous work has explored the integration of NVRAM into various architecture and system levels, an open question remains: do specific memory workload characteristics of scientific applications map well onto NVRAMs' features when used in a hybrid NVRAM-DRAM memory system? Furthermore, are there common classes of data structures used by scientific applications that should be frequently placed into NVRAM?In this paper, we analyze several mission-critical scientific applications in order to answer these questions. Specifically, we develop a binary instrumentation tool to statistically report memory access patterns in stack, heap, and global data. We carry out hardware simulation to study the impact of NVRAM for both memory power and system performance. Our study identifies many opportunities for using NVRAM for scientific applications. In two of our applications, 31% and 27% of the memory working sets are suitable for NVRAM. Our simulations suggest at least 27% possible power savings and reveal that the performance of some applications is insensitive to relatively long NVRAM write-access latencies.
Author Weikuan Yu
Marin, G.
Cira, C.
Zhuo Liu
McCurdy, C.
Dong Li
Vetter, J. S.
Author_xml – sequence: 1
  surname: Dong Li
  fullname: Dong Li
  email: lid1@ornl.gov
  organization: Oak Ridge Nat. Lab., Oak Ridge, TN, USA
– sequence: 2
  givenname: J. S.
  surname: Vetter
  fullname: Vetter, J. S.
  email: vetter@ornl.gov
  organization: Oak Ridge Nat. Lab., Oak Ridge, TN, USA
– sequence: 3
  givenname: G.
  surname: Marin
  fullname: Marin, G.
  email: maring@ornl.gov
  organization: Oak Ridge Nat. Lab., Oak Ridge, TN, USA
– sequence: 4
  givenname: C.
  surname: McCurdy
  fullname: McCurdy, C.
  email: cmcurdy@ornl.gov
  organization: Oak Ridge Nat. Lab., Oak Ridge, TN, USA
– sequence: 5
  givenname: C.
  surname: Cira
  fullname: Cira, C.
  email: cmc0031@auburn.edu
  organization: Auburn Univ., Auburn, IL, USA
– sequence: 6
  surname: Zhuo Liu
  fullname: Zhuo Liu
  email: zhuoliu@auburn.edu
  organization: Auburn Univ., Auburn, IL, USA
– sequence: 7
  surname: Weikuan Yu
  fullname: Weikuan Yu
  email: wkyu@auburn.edu
  organization: Auburn Univ., Auburn, IL, USA
BookMark eNotjE1PwjAcxmvERECOnrzsCxT7srbrERGVBIUE9Uq27l9THd3SlsR9exf19LzkeX4TNPKtB4SuKZlTSvTtene_288ZoWxe6DM006ogSmqRSyX0OZrQwXCileAjNKaCE8yIEpdoEuMnIYzwXI_R17oGn5ztnf_Itl3XhnTyLjmImW1DdtcnwIu6DhBjWTWQvbQev7dNmdwQnuHYhj5zPlt9pwBHwHtTDv3euF-oM9mi6xpnhnnr4xW6sGUTYfavU_T2sHpdPuHN9nG9XGywo0okrJkytdQFGMYtoTkpdGlA6EJaRiuQuq65NUIaaTioSlIrOK9qmnMoczOcpujmj-sA4NAFdyxDf5BMKk0o_wFOY11-
CODEN IEEPAD
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1109/IPDPS.2012.89
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Xplore POP ALL
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP All) 1998-Present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
EISBN 9780769546759
0769546757
EndPage 956
ExternalDocumentID 6267901
Genre orig-research
GroupedDBID 29O
6IE
6IF
6IH
6IK
6IL
6IN
AAJGR
AAWTH
ABLEC
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IEGSK
IPLJI
OCL
RIE
RIL
ID FETCH-LOGICAL-i175t-927cd698ec23f014089ace5986f21be69dd3fc56c6c3e7b61f533bd143ea4cc23
IEDL.DBID RIE
ISBN 1467309753
9781467309752
ISICitedReferencesCount 44
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000309131900083&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1530-2075
IngestDate Wed Aug 27 04:45:09 EDT 2025
IsPeerReviewed false
IsScholarly false
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i175t-927cd698ec23f014089ace5986f21be69dd3fc56c6c3e7b61f533bd143ea4cc23
PageCount 12
ParticipantIDs ieee_primary_6267901
PublicationCentury 2000
PublicationDate 2012-May
PublicationDateYYYYMMDD 2012-05-01
PublicationDate_xml – month: 05
  year: 2012
  text: 2012-May
PublicationDecade 2010
PublicationTitle 2012 IEEE 26th International Parallel and Distributed Processing Symposium
PublicationTitleAbbrev ipdps
PublicationYear 2012
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0020349
ssj0000781219
Score 1.745239
Snippet Future exascale systems face extreme power challenges. To improve power efficiency of future HPC systems, non-volatile memory (NVRAM) technologies are being...
SourceID ieee
SourceType Publisher
StartPage 945
SubjectTerms Instruments
Memory management
Nonvolatile memory
Phase change random access memory
Power demand
Resource management
Title Identifying Opportunities for Byte-Addressable Non-Volatile Memory in Extreme-Scale Scientific Applications
URI https://ieeexplore.ieee.org/document/6267901
WOSCitedRecordID wos000309131900083&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV07T8MwELZKxcBUoEW85YERt2mS2vFYoBUMlEgF1K1K_JAqUFKVFNF_z52TPgYWttiSo8h3jr-z7_uOkBswqfK05kyGwmOhtV0mFQSuNpVd6FGwpBJXbEKMRtFkIuMaud1wYYwxLvnMtPHR3eXrXC3xqKwD4FtIJGvtCSFKrtbmPAVFa3yUJquCLdRdKbVSPfAE0XOkLg7-jEzStdZT1fa34pudp_ghHmPKl9_G0u87JVfcjjNs_O9bD0lrS92j8WZTOiI1kx2Txrp2A62WcpN8lAxdx3KiL3NE4cvMqatSgLH0blUY1tfaVUhJPw0d5Rl7zzFxDhrPmJ67orOMDn4KPGBkYzC1Kd_uko9of-divEXehoPX-0dWFV5gM0ATBZO-UJrLyCg_sBiCRTJBulbErd9NDZdaB1b1uOIqMCLlXQugMdUAvUwSKhh0QupZnplTQuF_wLXlUiihQq5EKpMUxhuZysBLvPCMNHHupvNSW2NaTdv5390X5AAtUyYcXpJ6sViaK7KvvovZ1-LaOcQvDGmyKA
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3NT8IwFG-ImugJFYzf9uDRwr5o1yMqBCLMJaDhRra2S4hmIziM_Pf2dePj4MXb2qTL0ve6_l77fr-H0L02qbCkpIR7zCJektiECx24JjG3dY_QSyoyxSZYEPiTCQ8r6GHDhVFKmeQz1YBHc5cvM7GEo7KmBt-MA1lrv-V5jl2wtTYnKiBb44A4WRlugfJKoZZqaV9gLUProtqjgUu6Vnsq285WfrPZD5_DESR9OQ0o_r5TdMXsOd3q_772GNW35D0cbralE1RR6Smqrqs34HIx19BHwdE1PCf8OgccvkyNvirWQBY_rnJF2lKaGinxp8JBlpL3DFLndGMICborPEtx5yeHI0Yy0sZWxdtN-hFu71yN19FbtzN-6pGy9AKZaTyRE-4wISn3lXDcBIIwn0dA2PJp4tixolxKNxEtKqhwFYupnWjYGEsNvlTkCT3oDO2lWarOEdZ_BCoTyplgwqOCxTyK9XjFY-5akeVdoBrM3XReqGtMy2m7_Lv7Dh32xsPBdNAPXq7QEVipSD-8Rnv5Yqlu0IH4zmdfi1vjHL-Bt7Vv
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=2012+IEEE+26th+International+Parallel+and+Distributed+Processing+Symposium&rft.atitle=Identifying+Opportunities+for+Byte-Addressable+Non-Volatile+Memory+in+Extreme-Scale+Scientific+Applications&rft.au=Dong+Li&rft.au=Vetter%2C+J.+S.&rft.au=Marin%2C+G.&rft.au=McCurdy%2C+C.&rft.date=2012-05-01&rft.pub=IEEE&rft.isbn=9781467309752&rft.issn=1530-2075&rft.spage=945&rft.epage=956&rft_id=info:doi/10.1109%2FIPDPS.2012.89&rft.externalDocID=6267901
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1530-2075&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1530-2075&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1530-2075&client=summon