An analysis of an Alliant FX/8 system running Xylem (the operating system of the Cedar paralle superconducting experiment) is presented. Results for two distinct, real, scientific workloads executing on an Alliant FX/8 are discussed. A combination of user concurrency and system overhead measurements were taken for both workloads. Statistical cluster analysis is used to extract a state transition model to jointly characterize user concurrency and system overhead. A skewness factor is introduced and used to bring out the effects of unbalanced clustering when determining states with significant transitions.
|Original language||English (US)|
|Number of pages||8|
|Journal||Proceedings of the International Conference on Parallel Processing|
|State||Published - 1988|
ASJC Scopus subject areas
- Hardware and Architecture