Sensitivity of cache replacement policies

Abstract

The sensitivity of a cache replacement policy expresses to what extent the execution history may influence the number of cache hits and misses during program execution. We present an algorithm to compute the sensitivity of a replacement policy. We have implemented this algorithm in a tool called Relacs that can handle a large class of replacement policies including LRU, FIFO, PLRU, and MRU. Sensitivity properties obtained with Relacs demonstrate that the execution history can have a strong impact on the number of cache hits and misses if FIFO, PLRU, or MRU is used. A simple model of execution time is used to evaluate the impact of cache sensitivity on measured execution times. The model shows that measured execution times may strongly underestimate the worst-case execution time for FIFO, PLRU, and MRU.

References

1. Ahuja, R. K., Magnanti, T. L., and Orlin, J. B. 1993. Network Flows: Theory, Algorithms, and Applications. Prentice-Hall, Inc., Upper Saddle River, NJ.
2. Al-Zoubi, H., Milenkovic, A., and Milenkovic, M. 2004. Performance evaluation of cache replacement policies for the SPEC CPU2000 benchmark suite. In Proceedings of the ACM 42nd Annual Southeast Regional Conference. ACM Press, New York, NY, 267--272.
3. Berg, C. 2006. PLRU cache domino effects. In Proceedings of the 6th International Workshop on Worst-Case Execution Time (WCET) Analysis. F. Mueller, Ed., Internationales Begegnungs- und Forschungszentrum für Informatik (IBFI), Schloss Dagstuhl, Germany.
4. Bernat, G., Colin, A., and Petters, S. M. 2002. WCET analysis of probabilistic hard real-time systems. In Proceedings of the 23rd IEEE Real-Time Systems Symposium (RTSS'02). IEEE Computer Society, Los Alamitos, CA, 279.
5. Deverge, J.-F. and Puaut, I. 2005. Safe measurement-based WCET estimation. In Proceedings of the 5th International Workshop on Worst-Case Execution Time (WCET) Analysis. R. Wilhelm, Ed., Internationales Begegnungs- und Forschungszentrum für Informatik (IBFI), Schloss Dagstuhl, Germany.
6. Eklöv, D., Nikoleris, N., Black-Schaffer, D., and Hagersten, E. 2011. Cache pirating: Measuring the curse of the shared cache. In Proceedings of ICPP. 165--175.
7. Engblom, J. and Jonsson, B. 2002. Processor pipelines and their properties for static WCET analysis. In Proceedings of the 2nd International Conference on Embedded Software (EMSOFT'02). Springer-Verlag, 334--348.
8. Ferdinand, C., Heckmann, R., Langenbach, M., Martin, F., Schmidt, M., Theiling, H., Thesing, S., and Wilhelm, R. 2001. Reliable and precise WCET determination for a real-life processor. In Proceedings of the Embedded Software Workshop. Vol. 2211, 469--485.
9. Heckmann, R., Langenbach, M., Thesing, S., and Wilhelm, R. 2003. The influence of processor architecture on the design and the results of WCET tools. Proc. IEEE 91, 7, 1038--1054.
10. Hennessy, J. L. and Patterson, D. A. 1996. Computer Architecture: A Quantitative Approach 2nd Ed., Morgan Kaufmann.
11. Lawler, E. 1966. Optimal cycles in doubly weighted linear graphs. In Proceedings of the International Symposium on Theory of Graphs. 209--213.
12. Petters, S. M. 2002. Worst case execution time estimation for advanced processor architectures. Ph.D. thesis, Technische Universität München, Munich, Germany.
13. Puaut, I. and Decotigny, D. 2002. Low-complexity algorithms for static cache locking in multitasking hard real-time systems. In Proceedings of the 23rd IEEE Real-Time Systems Symposium (RTSS'02). IEEE Computer Society, Los Alamitos, CA, 114.
14. Reineke, J. 2008. Caches in WCET analysis. Ph.D. thesis, Universität des Saarlandes.
15. Reineke, J. and Grund, D. 2008. Relative competitive analysis of cache replacement policies. In Proceedings of the ACM SIGPLAN-SIGBED Conference on Languages, Compilers, and Tools for Embedded Systems. ACM, New York, NY, 51--60.
16. Reineke, J., Grund, D., Berg, C., and Wilhelm, R. 2007. Timing predictability of cache replacement policies. Real-Time Syst. 37, 2, 99--122.
17. Schlickling, M. and Pister, M. 2010. Semi-automatic derivation of timing models for WCET analysis. In Proceedings of the ACM SIGPLAN/SIGBED Conference on Languages, Compilers, and Tools for Embedded Systems. ACM, 67--76.
18. Silberschatz, A., Galvin, P. B., and Gagne, G. 2005. Operating System Concepts. Vol. 2nd. Addison-Wesley.
19. Theiling, H., Ferdinand, C., and Wilhelm, R. 2000. Fast and precise WCET prediction by separate cache and path analyses. Real-Time Syst. 18, 2/3.
20. Vera, X., Lisper, B., and Xue, J. 2003. Data cache locking for higher program predictability. SIGMETRICS Perform. Eval. Rev. 31, 1, 272--282.
21. Wenzel, I. 2006. Measurement-based timing analysis of superscalar processors. Ph.D. thesis, Technische Universität Wien, Institut für Technische Informatik.
22. Wilhelm, R., Engblom, J., Ermedahl, A., Holsti, N., Thesing, S., Whalley, D. B., Bernat, G., Ferdinand, C., Heckmann, R., Mitra, T., Mueller, F., Puaut, I., Puschner, P. P., Staschulat, J., and Stenström, P. 2008. The worst-case execution-time problem - overview of methods and survey of tools. ACM Trans. Embed. Comput. Syst. 7, 3.
23. Wilhelm, R., Grund, D., Reineke, J., Schlickling, M., Pister, M., and Ferdinand, C. 2009. Memory hierarchies, pipelines, and buses for future architectures in time-critical embedded systems. IEEE Trans. CAD Integrat. Circuits Syst. 28, 7, 966--978.

Index Terms

• General and reference

  • Cross-computing tools and techniques

    • Measurement

    • Metrics

• Hardware

  • Integrated circuits

    • Semiconductor memory

      • Dynamic memory