P. E. Crandall and M. J. Quinn. Problem decomposition in parallel networks. Technical report, Oregon State University, Department of Computer Science, March 1993. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Mesh Partitioning for Efficient Use of Distributed Systems - Jian Chen Valerie (2002) (3 citations) (Correct)
....one interconnection network. Some tools such as METIS, can produce partitions with unequal weights. However, none of these tools can take network performance into consideration in the partitioning process. For this reason, these tools are not applicable to distributed systems. Crandall and Quinn [7, 8, 9, 10, 11, 12] developed a partitioning advisory system for network of workstations. The advisory system has three built in partitioning methods (contiguous row, contiguous point, and block) Given information about the problem space, the machine speed, and the network, the advisory system provides ranking of ....
P. E. Crandall and M. J. Quinn. Problem decomposition in parallel networks. Technical report, Oregon State University, Department of Computer Science, March 1993.
PART: A Partitioning Tool for Efficient Use of Distributed.. - Valerie Taylor Jian (1997) (1 citation) (Correct)
....the speed. TOP DOMDEC is an interactive mesh partitioning tool. All these tools produce equal size partitions. These tools are applicable to systems with the same processors and one interconnection network. For this reason, these tools are not applicable to distributed systems. Crandall and Quinn [5, 6, 7, 8, 9, 10] developed a partitioning advisory system for network of workstations. The advisory system has three built in partitioning methods (contiguous row, contiguous point, block) Given information about the problem space, the machine speed, and the network, the advisory system provides ranking of the ....
P. E. Crandall and M. J. Quinn. Problem decomposition in parallel networks. Technical report, Oregon State University, Department of Computer Science, March 1993.
A Performance Monitoring System for Networked Parallel.. - Crandall, Nedeljkovic.. Self-citation (Crandall Quinn) (Correct)
....If the workload is not partitioned according to the abilities of the different processors, the faster processors will sit idle while the slower participants finish their computations. Static data partitioning for parallel processing in the heterogeneous workstation environment has been examined in [5, 6]. To effect static data partitioning, Dataparallel C relies on a data file that lists the available P0 = 5 P1 = 4 P2 = 4 P3 = 3 Relative Speeds P0 P1 P2 P3 Figure 5: An n Theta n homogeneous problem is partitioned across 4 processors with relative speeds (5,4,4,3) processors and their ....
Phyllis E. Crandall and Michael J. Quinn. Problem decomposition in parallel networks. Technical Report 93-80-05, Oregon State University, 1993.
Data Partitioning for Networked Parallel Processing - Crandall, Quinn (1993) (5 citations) Self-citation (Crandall Quinn) (Correct)
....the problem space for heterogeneous processors are binary recursive partitioning [5] Figure 2(b) and general quadrilateral decomposition [10] Figure 2(c) but these schemes often result in a larger number of communications than the other partitionings presented here. Crandall and Quinn [6] have shown that, in the worst case, binary recursive partitioning incurs 6p Gamma 4 communications and the general quadrilateral generates 8p Gamma 4 communications. 4 Comparisons and Conclusion The various methods examined in the previous section have widely differing requirements regarding ....
Phyllis E. Crandall and Michael J. Quinn. Problem decomposition in parallel networks. Technical Report 93-80-05, Oregon State University, 1993.
Block Data Decomposition for Partial-Homogeneous Parallel.. - Crandall, Quinn (1994) (2 citations) Self-citation (Crandall Quinn) (Correct)
....requirements, or the grid is sparse. Decomposition reduces to the problem of assigning equal computational effort to each processor. Block decomposition has specifically been addressed by [2, 3, 8, 9, 15, 18] and data decomposition across heterogeneous processors has been discussed in [4, 10, 16]. A number of systems to facilitate parallel processing on workstation clusters have recently appeared. PVM (Parallel Virtual Machine) which is freely available, supports parallel computing across a wide variety of architectures, including workstation networks [1, 12] Interprocessor ....
....General quadrilateral block decomposition with 9 processors. The grid is first partitioned as for homogeneous block and then distorted. P0 P1 P2 P3 P4 P5 P6 P7 P8 n n Figure 9: Recursive binary block decomposition with 9 processors. titioning generates 8p Gamma 4 messages during each iteration [10] assuming a toroidal problem space. Since the regularity of the partitioning has been destroyed, a lookup table showing the processor location of each grid point is required. The binary recursive decomposition scheme presented in [9] requires no more than 6p Gamma 4 messages per iteration and ....
Phyllis E. Crandall and Michael J. Quinn. Problem decomposition in parallel networks. Technical Report 93-80-05, Oregon State University, 1993.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC