Parallel Domain Decomposition and Load Balancing Using Space-Filling Curves

Authors:
Srinivas Aluru;Fatih E. Sevilgen
Affiliations:
-;-
Venue:
HIPC '97 Proceedings of the Fourth International Conference on High-Performance Computing
Year:
1997

Citing 0
Cited 8

Efficient Parallel Algorithms for Solvent Accessible Surface Area of Proteins

IEEE Transactions on Parallel and Distributed Systems
Locality aware dynamic load management for massively multiplayer games

Proceedings of the tenth ACM SIGPLAN symposium on Principles and practice of parallel programming
Scalable Parallel Octree Meshing for TeraScale Applications

SC '05 Proceedings of the 2005 ACM/IEEE conference on Supercomputing
From mesh generation to scientific visualization: an end-to-end approach to parallel supercomputing

Proceedings of the 2006 ACM/IEEE conference on Supercomputing
Comparison of message-passing and shared memory implementations of the GMRES method on MIMD computers

Scientific Programming
Mapping with Space Filling Surfaces

IEEE Transactions on Parallel and Distributed Systems
Fine-Grained Data Distribution Operations for Particle Codes

Proceedings of the 16th European PVM/MPI Users' Group Meeting on Recent Advances in Parallel Virtual Machine and Message Passing Interface
Load-balancing spatially located computations using rectangular partitions

Journal of Parallel and Distributed Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Partitioning techniques based on space-fillingcurves have received much recent attention due totheir low running time and good load balance characteristics. The basic idea underlying these methodsis to order the multidimensional data according toa space-filling curve and partition the resulting one-dimensional order. However, space-filling curves aredefined for points that lie on a uniform grid of a particular resolution. It is typically assumed that the coordinates of the points are representable using a fixednumber of bits, and the run-times of the algorithmsdepend upon the number of bits used.In this paper, we present a simple and efficient technique for ordering arbitrary and dynamic multidimensional data using space-filling curves and its application to parallel domain decomposition and load balancing. Our technique is based on a comparison routine that determines the relative position of two pointsin the order induced by a space-filling curve. The comparison routine could then be used in conjunction withany parallel sorting algorithm to eject parallel domaindecomposition.