Abstract | ||
---|---|---|
Existing file access predictors keep track of previous file access patterns and rely on a single heuristic to predict which of the previous successors to the file being currently accessed is the most likely to be accessed next. We present here a novel composite predictor that appliesmultiple heuristics to this selection problem. As a result, it can make use of specialized heuristics that can make very accurate predictions when access patterns are observed to meet their particular criteria. Simulation results involving a total of seven file access traces indicate that our predictor delivers more correct predictions and less inaccurate guesses than predictors relying on a single heuristic for selecting a successor. |
Year | DOI | Venue |
---|---|---|
2003 | 10.1109/MASS.2003.1194860 | IEEE Symposium on Mass Storage Systems |
Keywords | Field | DocType |
previous file access pattern,file access trace,accurate prediction,appliesmultiple heuristics,file access predictions,access pattern,multiple predictors,previous successor,existing file access predictor,specialized heuristics,single heuristic,novel composite predictor,space technology,pattern recognition,accuracy,associative memory,predictive models,system performance,memory management,operating systems,simulation | Heuristic,Content-addressable memory,Space technology,Successor cardinal,Computer science,Parallel computing,Memory management,Heuristics,Artificial intelligence,Machine learning | Conference |
ISSN | ISBN | Citations |
2160-195X | 0-7695-1914-8 | 10 |
PageRank | References | Authors |
0.59 | 11 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Gary A. S. Whittle | 1 | 10 | 0.59 |
Jehan-françois Pâris | 2 | 510 | 265.03 |
Ahmed Amer | 3 | 79 | 6.12 |
Darrell D. E. Long | 4 | 3111 | 536.40 |
Randal Burns | 5 | 1955 | 115.15 |