Gfursin: Created page with "Nearly 10 years ago, we found an interesting way of characterizing program behavior (particularly CPU/MEM bounds) by breaking code semantics, substituting array accesses with ..."

2013-10-23T08:20:43Z

Created page with "Nearly 10 years ago, we found an interesting way of characterizing program behavior (particularly CPU/MEM bounds) by breaking code semantics, substituting array accesses with ..."

New page

Nearly 10 years ago, we found an interesting way of characterizing program behavior (particularly CPU/MEM bounds) by breaking code semantics, substituting array accesses with scalar accesses and observing reaction. There we could predict CPU/MEM bounds 3 orders of magnitude faster than simulators and much more precise than hardware counters {{CMIND_PUB|976aac68c6834af6|Fur2004}} {{CMIND_PUB|b254c18c8794ba29|FOTP2004}}. Later, we used ''reactions to optimizations'' to find similar programs when predicting optimizations {{CMIND_PUB|9404f8911595b11b|FT2010}}.

We are gradually adding these techniques to cM and Alchemist, and plan to use it to analyze parallel applications for bottlenecks (removing or adding threads), etc - collaborations and help are very welcome!

Tools:CM:projects:reactions to changes - Revision history

Gfursin: Created page with "Nearly 10 years ago, we found an interesting way of characterizing program behavior (particularly CPU/MEM bounds) by breaking code semantics, substituting array accesses with ..."