Abstract

Performance bottleneck of parallel computer systems has mostly been I/O devices because of disparity between processor speed and I/O speed. Therefore I/O node placement strategy is required such that it can minimize the number of I/O nodes, I/O access time and I/O traffic in an interconnection network. In this paper, we propose an optimal distance-k embedding algorithm, and analyze its effect on system performance when this algorithm is applied to n x n torus architecture. We prove this algorithm is an efficient I/O node placement using software simulation. I/O node placement using the proposed algorithm shows the highest performance among other I/O node placements in all cases. It is because locations of I/O nodes are uniformly distributed in the whole network, resulting in reduced traffic in the intE'rconnection network.

참고문헌 (15)

1. Concurrent file system - Making highly parallel mass storage transparent , S. Cannon , Proceedings. of Supercomputing '89, St. Petersberg / v.,pp., 1989
   
2. Deadlock-free message routing in multiprocessor interconnection networks , W. J. Dally;C. L. Seitz , IEEE Transations on Computers / v.C-36,pp., 1987
   
3. Disk I/O in high-performance computing systems , D. Jensen , Ph.D Dissertation, University of Illinois / v.,pp., 1993
   
4. Optimal hot spot allocation on meshes for large-scale data-parallel algorithms , S. Y . Lee;C. M. Chen , IEEE Transactions on Parallel and Distributed Systems / v.6,pp.788-802, 1995
   
5. Design, analysis, and simulation of I/O architectures for hypercube multiprocessors , A. L. N. Reddy;P. Banerjee , IEEE Transactions on Parallel and Distributed Systems / v.1,pp.140-151, 1990
   
6. D. Reed.;R. Fujimoto , Multicomputer Networks / v.,pp., 1985
   
7. E. A. Brewer;C. N. Dellarocas , PROTEUS User Documentation, 545 Technology Square(0.5 edition) / v.,pp., 1992
   
8. K. Hwang , Advanced Computer Architecture / v.,pp., 1993
   
9. Parallel I/O subsystems for hypercube multicomputers , J. Ghosh;B. Agarwal , Proceedings of the Fifth International Parallel Processing Symposium / v.,pp.381-384, 1991
   
10. Lee distance and topological properties of k-ary n-cubes , B. Bose(et al.) , IEEE Transactions on Computers / v.44,pp.1021-1030, 1995
    
11. Resource placement in k-ary n-cubes , P. Ramanathan;S. Chalasani , '89 International Conference on Parallel Processing / v.,pp.Ⅱ-133-Ⅱ-140, 1992
    
12. I/O embeddings in hypercubes , A. L. N. Reddy;P. Banerjee;S. G. Abraham , Proc. 1988 Int. Conference on Parallel Processing / v.,pp.318-338, 1988
    
13. Performance evaluation of a paralle I/O subsystem for hypercube multicomputers , J. Ghosh(et al.) , Journal of Parallel and Distributed Computing / v.17,pp.96-106, 1993
    
14. The MIT alewife machine: a large-scale distributed-memory multiprocessor , A. Agarwal(et al.) , Scalable Shared-Memory Multiprocessors / v.,pp., 1991
    
15. E. R. Berlekamp , Algebraic coding theory / v.,pp., 1968
    

이 논문을 인용한 문헌 (0)

1. 이 논문을 인용한 문헌 없음