Please use this identifier to cite or link to this item: http://hdl.handle.net/20.500.12188/24144
DC FieldValueLanguage
dc.contributor.authorMitrevski, Pen_US
dc.contributor.authorGushev, Marjanen_US
dc.contributor.authorMishev, Anastasen_US
dc.date.accessioned2022-11-04T10:52:03Z-
dc.date.available2022-11-04T10:52:03Z-
dc.date.issued2000-06-13-
dc.identifier.urihttp://hdl.handle.net/20.500.12188/24144-
dc.description.abstractIn this article we review the concepts of branch, value and meniory prediction used in conjunction with control and Liata speculative execution in superscalm processors. Since the amount of available instruction level parallelism within a basic block is relatively small, control speculation techniques increase the number of candidate instructions for execution. Moreover, the integration of value and memory prediction in superscalar processors introduces a new kind of speculative execution. Data Jpeculation techniques allow the processor to execute instructions beyond the limit of true data RAW depemkncies. We identifi data speculation as a natural extension of control speculation and capture their similarities and diflerences. We also raise a new question: is the analytical mockelling approach really infeasible?en_US
dc.publisherIEEEen_US
dc.subjectsuperscalar processor, control and dzta speculation, branch prediction, value prediction, and memory predictionen_US
dc.titlePrediction and speculation techniques in ILPen_US
dc.typeProceeding articleen_US
dc.relation.conferenceITI 2000. Proceedings of the 22nd International Conference on Information Technology Interfaces (Cat. No. 00EX411)en_US
item.grantfulltextopen-
item.fulltextWith Fulltext-
crisitem.author.deptFaculty of Computer Science and Engineering-
crisitem.author.deptFaculty of Computer Science and Engineering-
Appears in Collections:Faculty of Computer Science and Engineering: Conference papers
Files in This Item:
File Description SizeFormat 
10.0000ieeexplore.ieee.org915824.pdf623.94 kBAdobe PDFView/Open
Show simple item record

Page view(s)

62
checked on Oct 11, 2024

Download(s)

24
checked on Oct 11, 2024

Google ScholarTM

Check


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