Kuei Yu Wang, Dept. of Comput. Sci., Purdue Univ., West Lafayette, IN, USA
D.C. Marinescu, Dept. of Comput. Sci., Purdue Univ., West Lafayette, IN, USA
We introduce a methodology for the analysis of the paging activity of parallel programs running on massively parallel systems. The methodology includes parallel program monitoring and the analysis of the collected data. We study the correlation of the paging activities of individual node programs in the SPMD execution mode and its effect on scheduling.
Index Terms:
program diagnostics; parallel programming; paged storage; scheduling; paging activity; individual node programs; SPMD execution mode; parallel programs; parallel program monitoring; scheduling
Citation:
Kuei Yu Wang, D.C. Marinescu, "Correlation of the paging activity of individual node programs in the SPMD execution mode," hicss, pp.61, 28th Hawaii International Conference on System Sciences (HICSS'95), 1995