Please use this identifier to cite or link to this item: http://dx.doi.org/10.14279/depositonce-8185.2
For citation please use:
Full metadata record
DC FieldValueLanguage
dc.contributor.authorLal, Sohan-
dc.contributor.authorLucas, Jan-
dc.contributor.authorJuurlink, Ben-
dc.date.accessioned2019-11-21T08:46:16Z-
dc.date.available2019-02-11T12:27:59Z-
dc.date.available2019-11-21T08:46:16Z-
dc.date.issued2019-05-16-
dc.identifier.isbn978-3-9819263-2-3-
dc.identifier.urihttps://depositonce.tu-berlin.de/handle/11303/9084.2-
dc.identifier.urihttp://dx.doi.org/10.14279/depositonce-8185.2-
dc.description.abstractMemory compression is a promising approach for reducing memory bandwidth requirements and increasing performance, however, memory compression techniques often result in a low effective compression ratio due to large memory access granularity (MAG) exhibited by GPUs. Our analysis of the distribution of compressed blocks shows that a significant percentage of blocks are compressed to a size that is only a few bytes above a multiple of MAG, but a whole burst is fetched from memory. These few extra bytes significantly reduce the compression ratio and the performance gain that otherwise could result from a higher raw compression ratio. To increase the effective compression ratio, we propose a novel MAG aware Selective Lossy Compression (SLC) technique for GPUs. The key idea of SLC is that when lossless compression yields a compressed size with few bytes above a multiple of MAG, we approximate these extra bytes such that the compressed size is a multiple of MAG. This way, SLC mostly retains the quality of a lossless compression and occasionally trades small accuracy for higher performance. We show a speedup of up to 35% normalized to a state-of-the-art lossless compression technique with a low loss in accuracy. Furthermore, average energy consumption and energy-delay- product are reduced by 8.3% and 17.5%, respectively.en
dc.description.sponsorshipEC/H2020/688759/EU/Low-Power Parallel Computing on GPUs 2/LPGPU2en
dc.language.isoenen
dc.relation.ispartof10.14279/depositonce-9156-
dc.rights.urihttp://rightsstatements.org/vocab/InC/1.0/en
dc.subject.ddc004 Datenverarbeitung; Informatiken
dc.subject.otherGPUen
dc.subject.othermemoryen
dc.subject.othercompressionen
dc.subject.othermemory access granularityen
dc.subject.otherMAGen
dc.subject.otherSelective Lossy Compressionen
dc.subject.otherSLCen
dc.titleSLC: Memory Access Granularity Aware Selective Lossy Compression for GPUsen
dc.typeConference Objecten
tub.accessrights.dnbdomainen
tub.publisher.universityorinstitutionTechnische Universität Berlinen
dc.identifier.eissn1558-1101-
dc.type.versionacceptedVersionen
dcterms.bibliographicCitation.doi10.23919/DATE.2019.8714810en
dcterms.bibliographicCitation.proceedingstitle2019 Design, Automation & Test in Europe Conference & Exhibition (DATE)en
dcterms.bibliographicCitation.originalpublisherplaceNew York, NYen
dcterms.bibliographicCitation.originalpublishernameInstitute of Electrical and Electronics Engineers (IEEE)en
Appears in Collections:FG Architektur eingebetteter Systeme » Publications

Files in This Item:
lal_etal_2019.pdf

Accepted manuscript

Format: Adobe PDF | Size: 252.46 kB
DownloadShow Preview
Thumbnail

Version History
Version Item Date Summary
2 10.14279/depositonce-8185.2 2019-11-21 09:38:32.317 add publisher metadata
1 10.14279/depositonce-8185 2019-02-11 13:27:59.0
Item Export Bar

Items in DepositOnce are protected by copyright, with all rights reserved, unless otherwise indicated.