This system will be undergoing maintenance Tuesday, December 6 from 9AM to 12PM CST.
CLUE: A Cluster Evaluation Tool

CLUE: A Cluster Evaluation Tool

Date: December 2006
Creator: Parker, Brandon S.
Description: Modern high performance computing is dependent on parallel processing systems. Most current benchmarks reveal only the high level computational throughput metrics, which may be sufficient for single processor systems, but can lead to a misrepresentation of true system capability for parallel systems. A new benchmark is therefore proposed. CLUE (Cluster Evaluator) uses a cellular automata algorithm to evaluate the scalability of parallel processing machines. The benchmark also uses algorithmic variations to evaluate individual system components' impact on the overall serial fraction and efficiency. CLUE is not a replacement for other performance-centric benchmarks, but rather shows the scalability of a system and provides metrics to reveal where one can improve overall performance. CLUE is a new benchmark which demonstrates a better comparison among different parallel systems than existing benchmarks and can diagnose where a particular parallel system can be optimized.
Contributing Partner: UNT Libraries
Evaluating Computer Program Performance on the CRAY-1

Evaluating Computer Program Performance on the CRAY-1

Date: January 1979
Creator: Rudsinski, Larry & Pieper, Gail W.
Description: The Advanced Scientific Computers Project of Argonne's Applied Mathematics Division has two objectives: to evaluate supercomputers and to determine their effect on Argonne's computing workload. Initial efforts have focused on the CRAY-1, which is the only advanced computer currently available. Users from seven Argonne divisions executed test programs on the CRAY and made performance comparisons with the IBM 370/195 at Argonne. This report describes these experiences and discusses various techniques for improving run times on the CRAY. Direct translations of code from scalar to vector processor reduced running times as much as two-fold, and this reduction will become more pronounced as the CRAY compiler is developed. Further improvement (two- to ten-fold) was realized by making minor code changes to facilitate compiler recognition of the parallel and vector structure within the programs. Finally, extensive rewriting of the FORTRAN code structure reduced execution times dramatically, in three cases by a factor of more than 20; and even greater reduction should be possible by changing algorithms within a production code. It is concluded that the CRAY-1 would be of great benefit to Argonne researchers. Existing codes could be modified with relative ease to run significantly faster than on the 370/195. More important, the ...
Contributing Partner: UNT Libraries Government Documents Department
Comparison of the CRAY X-MP-4, Fujitsu VP-200, and Hitachi S-810/20 : an Argonne Perspective

Comparison of the CRAY X-MP-4, Fujitsu VP-200, and Hitachi S-810/20 : an Argonne Perspective

Date: October 1985
Creator: Dongarra, J. J. & Hinds, Alan
Description: A set of programs, gathered from major Argonne computer users, was run on the current generation of supercomputers: the CRAY X-MP-4, Fujitsu VP-200, and Hitachi S-810/20. The results show that a single processor of a CRAY X-MP-4 is a consistently strong performer over a wide range of problems. The Fujitsu and Hitachi excel on highly vectorized programs and offer an attractive opportunity to sites with IBM-compatible computers.
Contributing Partner: UNT Libraries Government Documents Department