An on-line hot data identification for flash-based storage using sampling mechanism
Efficient hot and cold data identification in computer systems has been a fundamental issue. However, it has been least investigated. In this paper, we propose a novel on-line hot data identification scheme for flash-based storage named HotDataTrap. The main idea is to maintain a working set of pote...
Gespeichert in:
Veröffentlicht in: | Applied computing review : a publication of the Special Interest Group on Applied Computing 2013-03, Vol.13 (1), p.51-64 |
---|---|
Hauptverfasser: | , , , , , |
Format: | Artikel |
Sprache: | eng |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | 64 |
---|---|
container_issue | 1 |
container_start_page | 51 |
container_title | Applied computing review : a publication of the Special Interest Group on Applied Computing |
container_volume | 13 |
creator | Park, Dongchul Nam, Young Jin Debnath, Biplob Du, David H. C. Kim, Youngkyun Kim, Youngchul |
description | Efficient hot and cold data identification in computer systems has been a fundamental issue. However, it has been least investigated. In this paper, we propose a novel on-line hot data identification scheme for flash-based storage named HotDataTrap. The main idea is to maintain a working set of potential hot data items in a cache based on a sampling mechanism. This sampling-based scheme enables HotDataTrap to early discard some of the cold items so that it can reduce runtime overheads as well as a waste of memory spaces. Moreover, our two-level hash indexing scheme helps HotDataTrap directly look up a requested item in the cache and save a memory space further by exploiting spatial localities. Both our sampling approach and hierarchical hash indexing scheme empower HotDataTrap to precisely and efficiently identify hot data with a even less memory. Our extensive experiments with various realistic workloads demonstrate that our HotDataTrap outperforms the state-of-the-art scheme by an average of 335% and our two-level hash indexing scheme considerably improves further HotDataTrap performance up to 50.8%. |
doi_str_mv | 10.1145/2460136.2460141 |
format | Article |
fullrecord | <record><control><sourceid>crossref</sourceid><recordid>TN_cdi_crossref_primary_10_1145_2460136_2460141</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>10_1145_2460136_2460141</sourcerecordid><originalsourceid>FETCH-LOGICAL-c171t-70f3d1fa152120132e38505de59dfef4be3d39c0ab65d4aa423840487bb7bf3c3</originalsourceid><addsrcrecordid>eNotkDtPwzAURi0EEqUws_oPuPX1K8lYVTwqVWIA5ujGj9YocSo7DPx7Wsh0zvTp0yHkEfgKQOm1UIaDNKs_KrgiC2gkMA4Grs-udcNMA_qW3JXyxTnUYMSCvG8SHRPrY_L0OE7U4YQ0Op-mGKLFKY6JhjHT0GM5sg6Ld7RMY8aDp98lpgMtOJz6iwzeHjHFMtyTm4B98Q8zl-Tz-elj-8r2by-77WbPLFQwsYoH6SAgaAHifF14WWuundeNCz6ozksnG8uxM9opRCVkrbiqq66ruiCtXJL1_67NYynZh_aU44D5pwXeXpq0c5N2biJ_ARM_VGc</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>An on-line hot data identification for flash-based storage using sampling mechanism</title><source>ACM Digital Library Complete</source><creator>Park, Dongchul ; Nam, Young Jin ; Debnath, Biplob ; Du, David H. C. ; Kim, Youngkyun ; Kim, Youngchul</creator><creatorcontrib>Park, Dongchul ; Nam, Young Jin ; Debnath, Biplob ; Du, David H. C. ; Kim, Youngkyun ; Kim, Youngchul</creatorcontrib><description>Efficient hot and cold data identification in computer systems has been a fundamental issue. However, it has been least investigated. In this paper, we propose a novel on-line hot data identification scheme for flash-based storage named HotDataTrap. The main idea is to maintain a working set of potential hot data items in a cache based on a sampling mechanism. This sampling-based scheme enables HotDataTrap to early discard some of the cold items so that it can reduce runtime overheads as well as a waste of memory spaces. Moreover, our two-level hash indexing scheme helps HotDataTrap directly look up a requested item in the cache and save a memory space further by exploiting spatial localities. Both our sampling approach and hierarchical hash indexing scheme empower HotDataTrap to precisely and efficiently identify hot data with a even less memory. Our extensive experiments with various realistic workloads demonstrate that our HotDataTrap outperforms the state-of-the-art scheme by an average of 335% and our two-level hash indexing scheme considerably improves further HotDataTrap performance up to 50.8%.</description><identifier>ISSN: 1559-6915</identifier><identifier>EISSN: 1931-0161</identifier><identifier>DOI: 10.1145/2460136.2460141</identifier><language>eng</language><ispartof>Applied computing review : a publication of the Special Interest Group on Applied Computing, 2013-03, Vol.13 (1), p.51-64</ispartof><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c171t-70f3d1fa152120132e38505de59dfef4be3d39c0ab65d4aa423840487bb7bf3c3</citedby><cites>FETCH-LOGICAL-c171t-70f3d1fa152120132e38505de59dfef4be3d39c0ab65d4aa423840487bb7bf3c3</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,27924,27925</link.rule.ids></links><search><creatorcontrib>Park, Dongchul</creatorcontrib><creatorcontrib>Nam, Young Jin</creatorcontrib><creatorcontrib>Debnath, Biplob</creatorcontrib><creatorcontrib>Du, David H. C.</creatorcontrib><creatorcontrib>Kim, Youngkyun</creatorcontrib><creatorcontrib>Kim, Youngchul</creatorcontrib><title>An on-line hot data identification for flash-based storage using sampling mechanism</title><title>Applied computing review : a publication of the Special Interest Group on Applied Computing</title><description>Efficient hot and cold data identification in computer systems has been a fundamental issue. However, it has been least investigated. In this paper, we propose a novel on-line hot data identification scheme for flash-based storage named HotDataTrap. The main idea is to maintain a working set of potential hot data items in a cache based on a sampling mechanism. This sampling-based scheme enables HotDataTrap to early discard some of the cold items so that it can reduce runtime overheads as well as a waste of memory spaces. Moreover, our two-level hash indexing scheme helps HotDataTrap directly look up a requested item in the cache and save a memory space further by exploiting spatial localities. Both our sampling approach and hierarchical hash indexing scheme empower HotDataTrap to precisely and efficiently identify hot data with a even less memory. Our extensive experiments with various realistic workloads demonstrate that our HotDataTrap outperforms the state-of-the-art scheme by an average of 335% and our two-level hash indexing scheme considerably improves further HotDataTrap performance up to 50.8%.</description><issn>1559-6915</issn><issn>1931-0161</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2013</creationdate><recordtype>article</recordtype><recordid>eNotkDtPwzAURi0EEqUws_oPuPX1K8lYVTwqVWIA5ujGj9YocSo7DPx7Wsh0zvTp0yHkEfgKQOm1UIaDNKs_KrgiC2gkMA4Grs-udcNMA_qW3JXyxTnUYMSCvG8SHRPrY_L0OE7U4YQ0Op-mGKLFKY6JhjHT0GM5sg6Ld7RMY8aDp98lpgMtOJz6iwzeHjHFMtyTm4B98Q8zl-Tz-elj-8r2by-77WbPLFQwsYoH6SAgaAHifF14WWuundeNCz6ozksnG8uxM9opRCVkrbiqq66ruiCtXJL1_67NYynZh_aU44D5pwXeXpq0c5N2biJ_ARM_VGc</recordid><startdate>201303</startdate><enddate>201303</enddate><creator>Park, Dongchul</creator><creator>Nam, Young Jin</creator><creator>Debnath, Biplob</creator><creator>Du, David H. C.</creator><creator>Kim, Youngkyun</creator><creator>Kim, Youngchul</creator><scope>AAYXX</scope><scope>CITATION</scope></search><sort><creationdate>201303</creationdate><title>An on-line hot data identification for flash-based storage using sampling mechanism</title><author>Park, Dongchul ; Nam, Young Jin ; Debnath, Biplob ; Du, David H. C. ; Kim, Youngkyun ; Kim, Youngchul</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c171t-70f3d1fa152120132e38505de59dfef4be3d39c0ab65d4aa423840487bb7bf3c3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2013</creationdate><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Park, Dongchul</creatorcontrib><creatorcontrib>Nam, Young Jin</creatorcontrib><creatorcontrib>Debnath, Biplob</creatorcontrib><creatorcontrib>Du, David H. C.</creatorcontrib><creatorcontrib>Kim, Youngkyun</creatorcontrib><creatorcontrib>Kim, Youngchul</creatorcontrib><collection>CrossRef</collection><jtitle>Applied computing review : a publication of the Special Interest Group on Applied Computing</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Park, Dongchul</au><au>Nam, Young Jin</au><au>Debnath, Biplob</au><au>Du, David H. C.</au><au>Kim, Youngkyun</au><au>Kim, Youngchul</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>An on-line hot data identification for flash-based storage using sampling mechanism</atitle><jtitle>Applied computing review : a publication of the Special Interest Group on Applied Computing</jtitle><date>2013-03</date><risdate>2013</risdate><volume>13</volume><issue>1</issue><spage>51</spage><epage>64</epage><pages>51-64</pages><issn>1559-6915</issn><eissn>1931-0161</eissn><abstract>Efficient hot and cold data identification in computer systems has been a fundamental issue. However, it has been least investigated. In this paper, we propose a novel on-line hot data identification scheme for flash-based storage named HotDataTrap. The main idea is to maintain a working set of potential hot data items in a cache based on a sampling mechanism. This sampling-based scheme enables HotDataTrap to early discard some of the cold items so that it can reduce runtime overheads as well as a waste of memory spaces. Moreover, our two-level hash indexing scheme helps HotDataTrap directly look up a requested item in the cache and save a memory space further by exploiting spatial localities. Both our sampling approach and hierarchical hash indexing scheme empower HotDataTrap to precisely and efficiently identify hot data with a even less memory. Our extensive experiments with various realistic workloads demonstrate that our HotDataTrap outperforms the state-of-the-art scheme by an average of 335% and our two-level hash indexing scheme considerably improves further HotDataTrap performance up to 50.8%.</abstract><doi>10.1145/2460136.2460141</doi><tpages>14</tpages></addata></record> |
fulltext | fulltext |
identifier | ISSN: 1559-6915 |
ispartof | Applied computing review : a publication of the Special Interest Group on Applied Computing, 2013-03, Vol.13 (1), p.51-64 |
issn | 1559-6915 1931-0161 |
language | eng |
recordid | cdi_crossref_primary_10_1145_2460136_2460141 |
source | ACM Digital Library Complete |
title | An on-line hot data identification for flash-based storage using sampling mechanism |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-24T14%3A03%3A12IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-crossref&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=An%20on-line%20hot%20data%20identification%20for%20flash-based%20storage%20using%20sampling%20mechanism&rft.jtitle=Applied%20computing%20review%20:%20a%20publication%20of%20the%20Special%20Interest%20Group%20on%20Applied%20Computing&rft.au=Park,%20Dongchul&rft.date=2013-03&rft.volume=13&rft.issue=1&rft.spage=51&rft.epage=64&rft.pages=51-64&rft.issn=1559-6915&rft.eissn=1931-0161&rft_id=info:doi/10.1145/2460136.2460141&rft_dat=%3Ccrossref%3E10_1145_2460136_2460141%3C/crossref%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true |