Markov model prediction of I/O requests for scientific applications
- 22 June 2002
- proceedings article
- Published by Association for Computing Machinery (ACM)
- p. 147-155
- https://doi.org/10.1145/514191.514214
Abstract
Given the increasing performance disparity between processors and storage devices, exploiting knowledge of spatial and temporal I/O requests is critical to achieving high performance, particularly on parallel systems. Although perfect foreknowledge of I/O requests is rarely possible, even estimates of request patterns can potentially yield large performance gains. This paper evaluates Markov models to represent the spatial patterns of I/O requests in scientific codes. The paper also proposes three algorithms for I/O prefetching. Evaluation using I/O traces from scientific codes shows that highly accurate prediction of spatial access patterns, resulting in reduced execution times, is possible.Keywords
This publication has 10 references indexed in Scilit:
- ARIMA time series modeling and forecasting for adaptive I/O prefetchingPublished by Association for Computing Machinery (ACM) ,2001
- Diving deep: data-management and visualization strategies for adaptive mesh refinement simulationsComputing in Science & Engineering, 1999
- A Comparison of Logical and Physical Parallel I/o pAtternsThe International Journal of High Performance Computing Applications, 1998
- Lessons from characterizing the input/output behavior of parallel scientific applicationsPerformance Evaluation, 1998
- Input/output access pattern classification using hidden Markov modelsPublished by Association for Computing Machinery (ACM) ,1997
- A trace-driven comparison of algorithms for parallel prefetching and cachingPublished by Association for Computing Machinery (ACM) ,1996
- Optimal prefetching via data compressionJournal of the ACM, 1996
- Input/output characteristics of scalable parallel applicationsPublished by Association for Computing Machinery (ACM) ,1995
- Informed prefetching and cachingPublished by Association for Computing Machinery (ACM) ,1995
- Practical prefetching via data compressionPublished by Association for Computing Machinery (ACM) ,1993