MemSpy: analyzing memory system bottlenecks in programs
SIGMETRICS '92/PERFORMANCE '92 Proceedings of the 1992 ACM SIGMETRICS joint international conference on Measurement and modeling of computer systems
Hitting the memory wall: implications of the obvious
ACM SIGARCH Computer Architecture News
Continuous profiling: where have all the cycles gone?
ACM Transactions on Computer Systems (TOCS)
Tools for application-oriented performance tuning
ICS '01 Proceedings of the 15th international conference on Supercomputing
SIP: Performance Tuning through Source Code Interdependence
Euro-Par '02 Proceedings of the 8th International Euro-Par Conference on Parallel Processing
SIGMA: a simulator infrastructure to guide memory analysis
Proceedings of the 2002 ACM/IEEE conference on Supercomputing
The cache behaviour of large lazy functional programs on stock hardware
Proceedings of the 2002 workshop on Memory system performance
| Hi-index | 0.00 |
The increasing gap of processor and main memory performance underlines the need for cache-optimizations, especially on memory-intensive applications. Tools which are able to localize code regions with high cache miss ratio seem to be appropriate for access optimizations. However, a programmer often does not know what to do with the collected information. We try to improve this situation by providing cache reuse metrics which are supposed to give more precise hints on how to optimize memory access behavior. We enhanced the cache simulator Callgrind to give metrics on temporal and spatial cache utilization for a given memory block, relating this information to the code line where the block was loaded into cache. We show what is needed for hardware-supported measurement for such metrics, and give example code where the collected information directly points to optimization possibilities.