Load-sharing in heterogeneous systems via weighted factoring

Jeanette Schmidt~ R. N. Uma$ Joel Wein~ We consider the problem of scheduling a parallel loop with independent iterations on a network of heterogeneous workstations, and demonstrate the effectiveness of a variant of fa.toring, a scheduling policy originating in the context of shared address-space homogeneous multiprocessors. In the new scheme, weighted factoring, processors are dynamically assigned decreasing size chunks of iterations in proportion to their processing speeds. Through experiments on a network of SUN Spare workstations we show that weighted factoring significantly outperforms variants of a work-stea!ing load-balancing algorithm and on certain applications dramatically outperforms factoring as well. We then study weighted work assignment analytically, giving upper and lower bounds on its performance under the assumption that the processor iteration execution times can be modeled as weighted random variables. *Department of Computer Science,Polytechmc Umverslty, Brooklyn, NY, 11201. Researchsupported by ARPA/USAF under Grant no F30602-95-1-OO08and the New York State Science and Technology Foundation through Its center for Advanced Technology in Telecommunications Joel Wein wassupported in part by NSF Grant CCR-9211494, and Jeanette Schmidt m part by NSF grant CCR9305873. thummelQmono poly edu (Contact Author) *JpsC!qmcs4 poly.edu $ ruma@photon.poly .edu ~ wein@mem poly. edu. Permissionto makedigitallhard copiesof all or pastof thk material for personalor classroomuseis grantedwithout fee provided that the copies are not madeor dktributed for profit or commercialadvantage,the copyright notice, the title of the publication and its dateappear,and notice is given that copyright is by permissionof the ACM, Inc. To copy otherwise, to republish, to post on serversor to redistributeto lists, requiresspecific riersnissionand/or fee.