Issue No.04 - April (1995 vol.28)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/2.375175
Recent architecture and technology trends have led to a significant and increasing gap between processor and main memory speeds. Caches hide these latencies to some extent, but when cache misses are frequent, memory stalls can significantly degrade program execution time. This paper describes MemSpy, a performance monitoring system designed to help identify and fix program memory bottlenecks. The natural interrelationship between memory bottlenecks and program data structures motivates MemSpy's introduction of data oriented statistics for memory performance information. Furthermore, MemSpy's detailed statistics on the causes of cache misses are crucial for determining sources of memory bottlenecks.
Margaret Martonosi, Anoop Gupta, Thomas E. Anderson, "Tuning Memory Performance of Sequential and Parallel Programs", Computer, vol.28, no. 4, pp. 32-40, April 1995, doi:10.1109/2.375175