{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

quFiles_fast10 - quFiles The right le at the right time...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
USENIX Association FAST ’10: 8th USENIX Conference on File and Storage Technologies 1 quFiles: The right file at the right time KaushikVeeraraghavan # ,JasonFlinn # ,EdmundB.Nightingale andBrianNoble # UniversityofMichigan # MicrosoftResearch(Redmond) Abstract AquFileisaunifyingabstractionthatsimplifiesdata managementbyencapsulatingdifferentphysicalrepre- sentationsofthesamelogicaldata. SimilartoaquBit (quantumbit),theparticularrepresentationofthelogi- caldatadisplayedbyaquFileisnotdetermineduntilthe momentitisneeded. Therepresentationreturnedbya quFileisspecifiedbyadata-specificpolicythatcantake intoaccountcontextsuchastheapplicationrequesting thedata,thedeviceonwhichdataisaccessed,screen size,andbatterystatus. Wedemonstratethegeneral- ityofthequFileabstractionbyusingittoimplement sixcasestudies: resourcemanagement,copy-on-write versioning,dataredaction,resource-awaredirectories, application-awareadaptation,andplatform-specificen- coding. MostquFilepolicieswereexpressedusingless thanonehundredlinesofcode.Ourexperimentalresults showthat,withcachingandotherperformanceoptimiza- tions,quFilesaddlessthan1%overheadtoapplication- levelfilesystembenchmarks. 1 Introduction It has become increasingly common for new stor- age systems to implement context-awareadaptation , in which different representations of the same object are re- turned based on the context in which the object is ac- cessed. For instance, many systems transcode data to meet the screen size constraints of mobile devices [5, 12]. Others display reduced fidelity representations to meet constraints on resources such as network bandwidth [8, 27] and battery energy [11], display redacted representa- tionsofdatafileswhentheyareviewedatinsecureloca- tions [22, 42], and create different formats of multimedia data for diverse devices [29]. These systems, and many others, have been successful ataddressingspecificneedsforadaptingtherepresenta- tionofdatatofitagivencontext. However,theysuffer from several problems that inhibit their wide-scale adop- tion. First, building such systems is time-consuming. Most required several person-years to build a prototype; porting them to mainstream environments would be dif- ficult at best. Second, each system presentsa different abstraction and interface, so each has a learning curve. Third, these systems typically present only a single logi- calviewofdata,makingitdifficultforuserstopiercethe abstraction and explicitly choose different presentations. Whyaretheresomanysystemsthatsharethesame premise,yethavecompletelyseparateimplementations? The answer is that, as a community, we have failed to recognize that there is a fundamental abstraction that un- derlies all these systems. This simple abstraction is the ability to view different representations of the same log- ical data in different contexts.
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}