Code generation for general loops using methods from computational geometry

Andronikos, T; Ciorba, FM; Kamenopoulos, D; Theodoropoulos, P; Papakonstantinou, G

dc.contributor.author	Andronikos, T	en
dc.contributor.author	Ciorba, FM	en
dc.contributor.author	Kamenopoulos, D	en
dc.contributor.author	Theodoropoulos, P	en
dc.contributor.author	Papakonstantinou, G	en
dc.date.accessioned	2014-03-01T02:49:43Z
dc.date.available	2014-03-01T02:49:43Z
dc.date.issued	2004	en
dc.identifier.issn	10272658	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/34709
dc.relation.uri	http://www.scopus.com/inward/record.url?eid=2-s2.0-11844252569&partnerID=40&md5=59ba6c1d1006aa2bdb2b8626b01a878e	en
dc.subject	Automatic SPMD code generation	en
dc.subject	Dynamic scheduling	en
dc.subject	General loops	en
dc.subject	Message passing architectures	en
dc.subject.other	Algorithms	en
dc.subject.other	Iterative methods	en
dc.subject.other	Motion estimation	en
dc.subject.other	Parallel processing systems	en
dc.subject.other	Problem solving	en
dc.subject.other	Scheduling	en
dc.subject.other	Automatic SPMD code generation	en
dc.subject.other	Dynamic scheduling	en
dc.subject.other	General loops	en
dc.subject.other	Message passing architectures	en
dc.subject.other	Computational geometry	en
dc.title	Code generation for general loops using methods from computational geometry	en
heal.type	conferenceItem	en
heal.identifier.secondary	439-173	en
heal.publicationDate	2004	en
heal.abstract	This paper deals with general nested loops and proposes a novel dynamic scheduling technique. General loops contain complex loop bodies (consisting of arbitrary program statements, such as assignments, conditionals and repetitions) that exhibit uniform loop-carried dependencies. Therefore it is now possible to achieve efficient parallelization for a vast class of loops, mostly found in DSP, PDEs, signal and video coding. At the core of this technique lies a simple and efficient dynamic rule (SDS - Successive Dynamic Scheduling) for determining the next ready-to-be-executed iteration at runtime. The central idea is to schedule the iterations on-the-fly using SDS, along the optimal hyperplane (determined using the QuickHull algorithm). Furthermore, a tool (CRONUS/1) that implements this theory and automatically produces the SPMD parallel code for message passing architectures is presented. As a testing case study, the FSBM motion estimation algorithm (used in video coding standards, e.g., MPEG-2, H.261) was used. The experimental results validate the presented theory and corroborate the efficiency of the generated parallel code.	en
heal.journalName	Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems	en
dc.identifier.volume	16	en
dc.identifier.spage	348	en
dc.identifier.epage	353	en