Title
A Prefetch Taxonomy
Abstract
Abstract--The growing difference between processor and main memory cycle time demands the use of aggressive prefetch algorithms to reduce the effective memory access latency. However, prefetching can significantly increase memory traffic and unsuccessful prefetches may pollute the cache. Metrics such as coverage and accuracy result from a simplistic classification of individual prefetches as 驴good驴 or 驴bad.驴 They do not capture the full effect of each prefetch and, hence, do not accurately reflect the quality of the prefetch algorithm. Gross statistics such as changes in the number of misses, total traffic, and IPC are not attributable to individual prefetches. Such gross metrics are therefore useful only for ranking existing prefetch algorithms; they do not evaluate the effect of individual prefetches so that an algorithm might be tuned. In this paper, we introduce a new, accurate, and complete taxonomy, called the Prefetch Traffic and Miss Taxonomy (PTMT), for classifying each prefetch by precisely accounting for the difference in traffic and misses it generates, either directly or indirectly. We illustrate the use of PTMT by evaluating two data prefetch algorithms.
Year
DOI
Venue
2004
10.1109/TC.2004.1261824
IEEE Trans. Computers
Keywords
Field
DocType
unsuccessful prefetches,individual prefetches,memory traffic,ranking existing prefetch algorithm,prefetch taxonomy,total traffic,effective memory access latency,main memory cycle time,prefetch algorithm,aggressive prefetch algorithm,data prefetch algorithm,cycle time,algorithm design and analysis,statistics,ipc,cache memory,taxonomy,bandwidth
Ranking,Cache,Computer science,Latency (engineering),Parallel computing,Real-time computing,Storage management,Instruction prefetch
Journal
Volume
Issue
ISSN
53
2
0018-9340
Citations 
PageRank 
References 
22
1.03
17
Authors
3
Name
Order
Citations
PageRank
Vijayalakshmi Srinivasan1107783.50
Edward S. Davidson2922171.30
Gary S. Tyson357149.20