Workload Driven Index Defragmentation

Workload Driven Index Defragmentation

ID:40372566

大小:418.05 KB

页数:12页

时间:2019-08-01

Workload Driven Index Defragmentation _第1页
Workload Driven Index Defragmentation _第2页
Workload Driven Index Defragmentation _第3页
Workload Driven Index Defragmentation _第4页
Workload Driven Index Defragmentation _第5页
资源描述:

《Workload Driven Index Defragmentation 》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库

1、WorkloadDrivenIndexDefragmentationVivekNarasayya,ManojSyamalaMicrosoftResearchOneMicrosoftWay,Redmond,WA,USA.viveknar@microsoft.commanojsy@microsoft.comAbstract—Decisionsupportqueriesthatscanlargeindexescanincludingmeasuresofinternalandexternalfragmentation.Wesuffersigni

2、ficantdegradationinI/Operformanceduetoindexrefertotechniquesthatrelyonlyonthefragmentationfragmentation.DBAsrelyonrulesofthumbthatuseindexsizeinformationfordecidingwhichindexestodefragmentasdata-andfragmentationinformationtoaccomplishthetaskofdrivenapproaches.Sinceitisus

3、uallynotpossibletodecidingwhichindexestodefragment.However,therearetwodefragmentallindexeswithinatypicalbatchwindow(e.g.afundamentallimitationsthatmakethistaskchallenging.First,fewhoursatnight),DBAstodayuserulesofthumbtoselectdatabaseenginesofferlittlesupporttohelpestima

4、tetheimpactofdefragmentinganindexontheI/Operformanceofaquery.whichindexestodefragment,e.g.selectindexesthataremostSecond,defragmentationissupportedonlyatthegranularityoffragmented.anentireB+-Tree,whichcanbetoorestrictivesinceWeobservetwokeylimitationswiththestate-of-the-

5、artindefragmentationisanexpensiveoperation.Thispaperdescribesindexdefragmentation.First,thegranularityatwhichindextechniquesforaddressingtheabovelimitations.WealsostudydefragmentationissupportedisthefullB+-Tree,whichcantheproblemofselectingtheappropriateindexestodefragme

6、ntveryexpensiveforlargeindexes.Inmanycases,theforagivenworkload.WehaveimplementedourtechniquesinfragmentationmaynotbeuniformlydistributedacrosstheMicrosoftSQLServeranddevelopedatoolthatcanprovideB+Tree.Forexample,consideraclusteredindexontheappropriateindexdefragmentatio

7、nrecommendationstoDBAs.OrderDatecolumnofalargefacttablethatstoresorderWeevaluatetheeffectivenessoftheproposedtechniquesonseveralrealandsyntheticdatabases.information.Asnewdataisinsertedintothefacttable,theB+-Treegetsfragmented.However,itisoftenthecasethatI.INTRODUCTIONma

8、nyofthequeriesreferencingonlyrecentdata(e.g.lastDecisionsupportqueriesoftenrequirescansoflargemonthorla

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。