Show simple item record

dc.contributor.authorZhang, Xiaochun
dc.contributor.authorJones, Timothy M
dc.contributor.authorCampanoni, Simone
dc.date.accessioned2021-10-15T23:31:14Z
dc.date.available2021-10-15T23:31:14Z
dc.identifier.urihttps://www.repository.cam.ac.uk/handle/1810/329445
dc.description.abstractAutomatic parallelizing compilers are often constrained in their transformations because they must conservatively respect data dependences within the program. Developers, on the other hand, often take advantage of domain-specific knowledge to apply transformations that modify data dependences but respect the application’s semantics. This creates a semantic gap between the parallelism extracted automatically by compilers and manually by developers. Although prior work has proposed programming language extensions to close this semantic gap, their relative contribution is unclear and it is uncertain whether compilers can actually achieve the same performance as manually parallelized code when using them. We quantify this semantic gap in a set of sequential and parallel programs and leverage these existing programming-language extensions to empirically measure the impact of closing it for an automatic parallelizing compiler. This lets us achieve an average speedup of 12.6× on an Intel-based 28-core machine, matching the speedup obtained by the manually parallelized code. Further, we apply these extensions to widely used sequential system tools, obtaining 7.1× speedup on the same system.
dc.publisherIEEE
dc.rightsAll rights reserved
dc.titleQuantifying the Semantic Gap Between Serial and Parallel Programming
dc.typeConference Object
dc.identifier.doi10.17863/CAM.76893
dcterms.dateAccepted2021-09-03
rioxxterms.versionofrecord10.17863/CAM.76893
rioxxterms.versionAM
rioxxterms.licenseref.urihttp://www.rioxx.net/licenses/all-rights-reserved
rioxxterms.licenseref.startdate2021-09-03
dc.contributor.orcidJones, Timothy M. [0000-0002-4114-7661]
rioxxterms.typeConference Paper/Proceeding/Abstract
pubs.funder-project-idEngineering and Physical Sciences Research Council (EP/K026399/1)
pubs.funder-project-idEngineering and Physical Sciences Research Council (EP/P020011/1)
datacite.issupplementedby.urlhttps://doi.org/10.17863/CAM.76224
pubs.conference-nameIEEE International Symposium on Workload Characterization
pubs.conference-start-date2021-11-07
cam.orpheus.counter34*
pubs.conference-finish-date2021-11-09
rioxxterms.freetoread.startdate2024-10-15


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record