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.
- R. Aydt. The Pablo Self-Defining Data Format. Technical report, Department of Computer Science, University of Illinois, April 1994.Google Scholar
- C. Bona, J. Masso, E. Seidel, and P. Walker. Three Dimensional Numerical Relativity with a Hyperbolic Formulation. submitted to Physics Reviews D, 1998.Google Scholar
- T. Cortes and J. Labarta. Linear Aggressive Prefetching: a Aay to Increase Performance of Cooperative Caches. In Proceedings of the 13th International Parallel Processing Symposium, April 1999. Google ScholarDigital Library
- P. E. Crandall, R. A. Aydt, A. A. Chien, and D. A. Reed. Characterization of a Suite of Input/Output Intensive Applications. In Proceedings of Supercomputing '95, Dec. 1995.Google Scholar
- P. E. Crandall, R. A. Aydt, A. A. Chien, and D. A. Reed. Input/Output Characteristics of Scalable Parallel Applications. In Proceedings of Supercomputing '95, Dec. 1995. Google ScholarDigital Library
- K. M. Curewitz, P. Krishnan, and J. S. Vitter. Practical Prefetching via Data Compression. In Proceedings of the 1993 ACM-SIGMOD Conference on Management of Data, pages 257--266, 1993. Google ScholarDigital Library
- R. Jain. The Art of Computer Systems Performance Analysis. Wiley, 1991.Google Scholar
- T. Kimbrel, A. Tomkins, R. Patterson, B. Bershad, P. Cao, E. Felten, G. Gibson, A. Karlin, and K. Li. A Trace-Driven Comparison of Algorithms for Parallel Prefetching and Caching. In 2nd USENIX Symposium on Operating Systems Design and Implementation, pages 19--34, October 1996. Google ScholarDigital Library
- T. M. Madhyastha. Automatic Classification of Input/Output Access Patterns. PhD thesis, Department of Computer Science, University of Illinois at Urbana-Champaign, 1997. Google ScholarDigital Library
- T. M. Madhyastha and D. A. Reed. Input/Output Access Pattern Classification Using Hidden Markov Models. In Proceedings of the Fifth Workshop on Input/Output in Parallel and Distributed Systems, pages 57--67, Nov. 1997. Google ScholarDigital Library
- M. L. Norman, J. Shalf, S. Levy, and G. Daues. Diving deep: Data-management and visualization strategies for adaptive mesh refinement simulations. Computing in Science and Engineering, 1(4):36--47, Jul/Aug 1999. Google ScholarDigital Library
- R. H. Patterson, G. A. Gibson, E. Ginting, D. Stodolsky, and J. Zelenka. Informed prefetching and caching. In Proceedings of the Fifteenth ACM Symposium on Operating Systems Principles, pages 79--95, December 1995. Google ScholarDigital Library
- D. A. Reed, editor. Scalable Input/Output: Achieving System Balance. MIT Press, in prep. Google ScholarDigital Library
- H. Simitci and D. A. Reed. A Comparison of Logical and Physical Parallel I/O Patterns. International Journal of High Performance Computing Applications, 12(3):364--380, 1998.Google ScholarDigital Library
- E. Smirni, R. A. Aydt, A. A. Chien, and D. A. Reed. I/O Requirements of Scientific Applications: An Evolutionary View. In Proceedings of the Fifth IEEE International Symposium on High-Performance Distributed Computing, pages 49--59, Aug. 1996. Google ScholarDigital Library
- E. Smirni, C. L. Elford, and D. A. Reed. Performance Modeling of a Parallel I/O System: An Application Driven Approach. In Proceedings of the Eighth SIAM Conference on Parallel Processing for Scientific Computing, Mar. 1997.Google Scholar
- E. Smirni and D. A. Reed. Workload Characterization of Input/Output Intensive Parallel Applications. In Proceedings of the 9th International Conference on Modelling Techniques and Tools for Computer Performance Evaluation, pages 169--180. Springer-Verlag, June 1997. Google ScholarDigital Library
- E. Smirni and D. A. Reed. Lessons from Characterizing the Input/Output Behavior of Parallel Scientific Applications. Performance Evaluation: An International Journal, 33(1):27--44, June 1998. Google ScholarDigital Library
- N. Tran and D. A. Reed. ARIMA Time Series Modeling and Forecasting for Adaptive I/O Prefetching. In Proceedings of the 15th ACM International Conference on Supercomputing, pages 473--485, June 2001. Google ScholarDigital Library
- N. N. Tran. Automatic ARIMA Time Series Modeling and Forecasting for Adaptive Input/Output Prefetching. PhD thesis, University of Illinois at Urbana-Champaign, Dec. 2001. Google ScholarDigital Library
- V. Vellanki and A. L. Chervenak. Prefetching Without Hints: A Cost-Benefit Analysis for Predicted Accesses. Technical Report GIT-CC-99-07, Georgia Tech, 1999.Google Scholar
- J. S. Vitter and P. Krishnan. Optimal prefetching via data compression. Journal of the ACM, 43(5):771--793, September 1996. Google ScholarDigital Library
Index Terms
- Markov model prediction of I/O requests for scientific applications
Recommendations
S-CAVE: effective SSD caching to improve virtual machine storage performance
PACT '13: Proceedings of the 22nd international conference on Parallel architectures and compilation techniquesA unique challenge for SSD storage caching management in a virtual machine (VM) environment is to accomplish the dual objectives: maximizing utilization of shared SSD cache devices and ensuring performance isolation among VMs. In this paper, we present ...
Data value prefetching method based on Markov model
ICCOMP'06: Proceedings of the 10th WSEAS international conference on ComputersData value prefetching at the beginning of the pipeline is one approach to overcoming data dependences and improving instruction-level parallelism. In this paper, a Markov model used on data value prefetching is presented. The paper investigates ...
Comments