Input/output access pattern classification using hidden Markov models

Input/output performance on current parallel file systems is sensitive to a good match of application access pattern to file system capabilities. Automaticiuput/output access classification can determine application access patterns at execution time, guiding adaptive file system policies. In this paper we examine a new method for access pattern classification that uses hidden Markov models, trained on access patterns from previous executions, to create a probabilistic model of input/output accesses. We compare this approach to a neural network classification &n-rework, presenting performance results from parallel and sequential benchmarks and applications.

[1]  S. Wittevrongel,et al.  Queueing systems , 2019, Autom..

[2]  K. Korner,et al.  Intelligent caching for remote file service , 1990, Proceedings.,10th International Conference on Distributed Computing Systems.

[3]  Andrew A. Chien,et al.  I/O requirements of scientific applications: an evolutionary view , 1996, Proceedings of 5th IEEE International Symposium on High Performance Distributed Computing.

[4]  Andrew S. Grimshaw,et al.  ELFS: object-oriented extensible file systems , 1991, [1991] Proceedings of the First International Conference on Parallel and Distributed Information Systems.

[5]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[6]  T.M. Madhyastha,et al.  Intelligent, adaptive file system policy selection , 1996, Proceedings of 6th Symposium on the Frontiers of Massively Parallel Computation (Frontiers '96).

[7]  Hui Lei,et al.  An analytical approach to file prefetching , 1997 .

[8]  Thomas M. Kroeger,et al.  Predicting file system actions from prior events , 1996 .

[9]  Peter F. Corbett,et al.  Proposal for a Common Parallel File System Programming Interface 1.0 , 1996, WWW 1996.

[10]  Jim Zelenka,et al.  Informed prefetching and caching , 1995, SOSP.

[11]  Stanley B. Zdonik,et al.  Fido: A Cache That Learns to Fetch , 1991, VLDB.

[12]  George Em Karniadakis,et al.  Unstructured spectral element methods for simulation of turbulent flows , 1995 .

[13]  Aron Kuppermann,et al.  The quantitative prediction and lifetime of a pronounced reactive scattering resonance , 1995 .

[14]  Daniel A. Reed,et al.  Automatic classification of input/output access patterns , 1997 .

[15]  Jim Griffioen,et al.  Reducing File System Latency using a Predictive Approach , 1994, USENIX Summer.

[16]  Andrew Tomkins,et al.  Informed multi-process prefetching and caching , 1997, SIGMETRICS '97.

[17]  Andrew A. Chien,et al.  PPFS: a high performance portable parallel file system , 1995, ICS '95.

[18]  Eugene Charniak,et al.  Statistical language learning , 1997 .

[19]  Steven A. Cuccaro,et al.  Quantum chemical reaction dynamics on a highly parallel supercomputer , 1991 .

[20]  Padhraic Smyth,et al.  Clustering Sequences with Hidden Markov Models , 1996, NIPS.