The Community for Technology Leaders
Green Image
<p><b>Abstract</b>—We discuss Inverse Spacefilling Partitioning (ISP), a partitioning strategy for non-uniform scientific computations running on distributed memory MIMD parallel computers. We consider the case of a dynamic workload distributed on a uniform mesh, and compare ISP against Orthogonal Recursive Bisection (ORB) and a Median of Medians variant of ORB, ORB-MM. We present two results. First, ISP and ORB-MM are superior to ORB in rendering balanced workloads—because they are more fine-grained—and incur communication overheads that are comparable to ORB. Second, ISP is more attractive than ORB-MM from a software engineering standpoint because it avoids elaborate bookkeeping. Whereas ISP partitionings can be described succinctly as logically contiguous segments of the line, ORB-MM's partitionings are inherently unstructured. We describe the general <it>d</it>-dimensional ISP algorithm and report empirical results with two- and three-dimensional, non-hierarchical particle methods.</p>
Dynamic load balancing, distributed memory parallel computers, spacefilling curve, non-uniform computations, irregular data partitioning, particle methods, performance.
John R. Pilkington, Scott B. Baden, "Dynamic Partitioning of Non-Uniform Structured Workloads with Spacefilling Curves", IEEE Transactions on Parallel & Distributed Systems, vol. 7, no. , pp. 288-300, March 1996, doi:10.1109/71.491582
92 ms
(Ver 3.3 (11022016))