Profiling Macro Data Flow Graphs for Parallel Implementation of FDTD Computations

Author

Smyk, Adam ; Tudruj, Marek

Author_Institution

Polish-Japanese Inst. of Inf. Technol., Warsaw

fYear

2005

fDate

4-6 July 2005

Firstpage

121

Lastpage

130

Abstract

In this paper, we present methodology, which enables designing and profiling macro data flow graphs that represent computation and communication patterns for the FDTD (finite difference time domain) problem in irregular computational areas. Optimized macro dataflow graphs (MDFG) for FDTD computations are generated in three main phases: generation of initial MDFG based on wave propagation area partitioning, MDFG nodes merging with load balancing to obtain given number of macro nodes and communication optimization to minimize and balance internode data transmissions. The computation efficiency for several communication systems (MPI, RDMA RB, SHMEM) is discussed. Relations between communication optimization algorithms and overall FDTD computation efficiency are shown. Experimental results obtained by simulation are presented

Keywords

data flow graphs; finite difference time-domain analysis; mathematics computing; message passing; parallel processing; resource allocation; wave propagation; MPI; RDMA RB; SHMEM; communication optimization; finite difference time domain; load balancing; macro data flow graphs; parallel FDTD computations; wave propagation area partitioning; Concurrent computing; Data communication; Data flow computing; Design methodology; Finite difference methods; Flow graphs; Load management; Merging; Partitioning algorithms; Time domain analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Parallel and Distributed Computing, 2005. ISPDC 2005. The 4th International Symposium on

Conference_Location

Lille

Print_ISBN

0-7695-2434-6

Type

conf

DOI

10.1109/ISPDC.2005.40

Filename

1609962