Evaluation of hardware-based stride and sequential prefetching in shared-memory multiprocessors
- 1 April 1996
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 7 (4) , 385-398
- https://doi.org/10.1109/71.494633
Abstract
We study the efficiency of previously proposed stride and sequential prefetching-two promising hardware-based prefetching schemes to reduce read-miss penalties in shared-memory multiprocessors. Although stride accesses dominate in four out of six of the applications we study, we find that sequential prefetching does as well as and in same cases even better than stride prefetching for five applications. This is because 1) most strides are shorter than the block size (we assume 32 byte blocks), which means that sequential prefetching is as effective for these stride accesses, and 2) sequential prefetching also exploits the locality of read misses with nonstride accesses. However, since stride prefetching in general results in fewer useless prefetches, it offers the extra advantage of consuming less memory-system bandwidth.Keywords
This publication has 17 references indexed in Scilit:
- The Cachemire Test Bench A Flexible And Effective Approach For Simulation Of MultiprocessorsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- Combined performance gains of simple cache protocol extensionsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- A performance study of software and hardware data prefetching schemesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Effectiveness of hardware-based stride and sequential prefetching in shared-memory multiprocessorsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- A Preliminary Evaluation of Cache-Miss-Initiated Prefetching Techniques in Scalable MultiprocessorsPublished by Defense Technical Information Center (DTIC) ,1994
- Prefetch unit for vector operations on scalar computersACM SIGARCH Computer Architecture News, 1992
- SPLASHACM SIGARCH Computer Architecture News, 1992
- An effective on-chip preloading scheme to reduce data access penaltyPublished by Association for Computing Machinery (ACM) ,1991
- A survey of cache coherence schemes for multiprocessorsComputer, 1990
- A New Solution to Coherence Problems in Multicache SystemsIEEE Transactions on Computers, 1978