GPUfs: The Case for Operating System Services on GPUs.
The article focuses on making core operating system abstractions available to graphics processing unit (GPU) code and the lessons learned in building GPUf file system layer. Topics discussed include how GPU have acquired increasingly general-purpose parallel processing capabilities, the traditional...
| Published in: | Communications of the ACM Vol. 57; no. 12; pp. 68 - 80 |
|---|---|
| Main Authors: | , , |
| Format: | Article |
| Published: |
Association for Computing Machinery
Dec2014
|
| Subjects: | |
| Online Access: | View this record in EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=hlh&AN=99744610&site=ehost-live header: @attributes: shortDbName: hlh uiTerm: 99744610 longDbName: Humanities International Complete uiTag: AN controlInfo: bkinfo: jinfo: jid: 00010782 ACM jtl: Communications of the ACM issn: 00010782 maglogo: N pubinfo: dt: Dec2014 vid: 57 iid: 12 pid: 68 pub: Association for Computing Machinery artinfo: ui: 99744610 10.1145/2656206 ppf: 68 ppct: 12 formats: tig: atl: GPUfs: The Case for Operating System Services on GPUs. aug: au: SILBERSTEIN, MARK FORD, BRYAN WITCHEL, EMMETT affil: Assistant professor, Department of Electrical Engineering, The Technion -- Israel Institute of Technology, Haifa, Israel Associate professor, Department of Computer Science, Yale University, New Haven, CT Associate professor, Department of Computer Science, University of Texas, Austin su: Graphics processing units Computer operating systems Parallel processing Computer software Computer input-output equipment Computer systems sug: subj: Graphics processing units Computer operating systems Parallel processing Computer software Computer input-output equipment Computer systems ab: The article focuses on making core operating system abstractions available to graphics processing unit (GPU) code and the lessons learned in building GPUf file system layer. Topics discussed include how GPU have acquired increasingly general-purpose parallel processing capabilities, the traditional CPU-centric programming model and how GPUfs demonstrates the productivity and performance benefits of allowing GPU to guide data flow in a system. An overview of the GPU software/hardware model is given. pubtype: Periodical doctype: Article src: R language: English refInfo: copyright: @attributes: flag: Y dt: @attributes: year: 2014 holdings: @attributes: islocal: N |
|---|