Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
An E cient Mean Field Annealing Formulation for Mapping Unstructured Domains to
 

Summary: An E cient Mean Field Annealing Formulation
for Mapping Unstructured Domains to
Hypercubes?
Cevdet Aykanat and _Ismail Haritaoglu
Computer Engineering Department, Bilkent University, Ankara, Turkey
aykanat@cs.bilkent.edu.tr
Abstract. We propose an e cient MFA formulation for mapping un-
structured domains to hypercube-connected distributed-memory archi-
tectures. In the general MFA formulation, N P spin variables are main-
tained and an individual MFA iteration requires (davgP +P2
) time for
the mapping of a sparse domain graph with N vertices and average vertex
degree of davg to a parallel architecture with P processors. The proposed
hypercube-speci c MFA formulation asymptotically reduces the number
of spin variables and the computational complexity of an individual MFA
iteration to Nlg2P and (davglg2P +Plg2P), respectively, by exploiting
the topological properties of hypercubes.
1 Introduction
Parallelization schemes for many applications on distributed memory architec-
tures are characterized by the mappingof the problemdomainto processors with

  

Source: Aykanat, Cevdet - Department of Computer Engineering, Bilkent University

 

Collections: Computer Technologies and Information Sciences