The non-critical buffer: using load latency tolerance to improve data cache efficiency
- 20 January 2003
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Abstract
Data cache performance is critical to overall processor performance as the latency gap between CPU core and main memory increases. Studies have shown that some loads have latency demands that allow them to be serviced from slower portions of memory, thus allowing more critical data to be kept in higher levels of the cache. We provide a strategy for identifying this latency-tolerant data at runtime and, using simple heuristics, keep it out of the main cache and place it instead in a small, parallel, associative buffer. Using such a "Non-Critical Buffer" dramatically improves the hit rate for more critical data, and leads to a performance improvement comparable to or better than other traditional cache improvement schemes. IPC improvements of over 4% are seen for some benchmarks.Keywords
This publication has 11 references indexed in Scilit:
- Column-associative Caches: A Technique For Reducing The Miss Rate Of Direct-mapped CachesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- Reducing conflicts in direct-mapped caches with a temporality-based designPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Improving direct-mapped cache performance by the addition of a small fully-associative cache and prefetch buffersPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Load latency tolerance in dynamically scheduled processorsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Design and performance evaluation of a cache assist to implement selective cachingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- A comparison of software code reordering and victim buffersACM SIGARCH Computer Architecture News, 1999
- Power and performance tradeoffs using various caching strategiesPublished by Association for Computing Machinery (ACM) ,1998
- Managing data caches using selective cache line replacementInternational Journal of Parallel Programming, 1997
- Cache performance of the SPEC92 benchmark suiteIEEE Micro, 1993
- Cache performance of operating system and multiprogramming workloadsACM Transactions on Computer Systems, 1988