Chaos in computer performance

Hugues  Berry; Daniel  Gracia Pérez; Olivier  Temam

doi:10.1063/1.2159147

ABSTRACT

Modern computer microprocessors are composed of hundreds of millions of transistors that interact through intricate protocols. Their performance during program execution may be highly variable and present aperiodic oscillations. In this paper, we apply current nonlinear time series analysis techniques to the performances of modern microprocessors during the execution of prototypical programs. Our results present pieces of evidence strongly supporting that the high variability of the performance dynamics during the execution of several programs display low-dimensional deterministic chaos, with sensitivity to initial conditions comparable to textbook models. Taken together, these results show that the instantaneous performances of modern microprocessors constitute a complex (or at least complicated) system and would benefit from analysis with modern tools of nonlinear and complexity science.

REFERENCES
Section:

1. G. E. Moore, Electronics 38, 114 (1965). Google Scholar
2. M. Stephenson and S. Amarasinghe, in Proc. International Symposium on Code Generation and Optimization (CGO), 2005. Google Scholar
3. P. Kulkarni, S. Hines, J. Hiser, D. Whalley, J. Davidson, and D. Jones, in Proc. ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI), 2004. Google Scholar
4. G. Fursin, M. O’Boyle, and P. Knijnenburg, in Proc. Languages and Compilers for Parallel Computers (LCPC), 2002, pp. 305–315. Google Scholar
5. E. Duesterwald, C. Cascaval, and S. Dwarkadas, in Proc. 12th International Conference on Parallel Architectures and Compilation Techniques (PACT’03), New Orleans, Louisiana, 2003, pp. 220–231. Google Scholar
6. M. Annavaram, R. Rakvic, M. Polito, J. -Y. Bouguet, R. Hankins, and B. Davies, in Proc. 37th Annual IEEE ACM International Symposium on Microarchitecture (MICRO-37), Portland, OR, 2004, pp. 93–104. Google Scholar
7. N. T. Slingerland and A. J. Smith, in ICS ’01: Proceedings of the 15th International Conference on Supercomputing (ACM Press, New York, NY, 2001), pp. 204–217, ISBN 1-58113-410-X. Google ScholarCrossref
8. M. D. Lam, E. E. Rothberg, and M. E. Wolf, in ASPLOS-IV: Proceedings of the 4th International Conference on Architectural Support for Programming Languages and Operating Systems (ACM Press, New York, NY, 1991), pp. 63–74, ISBN 0-89791-380-9. Google ScholarCrossref
9. S. Coleman and K. McKinley, in Proc. PLDI, 1995, pp. 279–290. Google ScholarCrossref
10. I. Gluhovsky and B. O’Krafka, ACM Trans. Comput. Syst. 23, 111 (2005). Google ScholarCrossref
11. J. Abella, A. Gonzalez, J. Llosa, and X. Vera, “Near-optimal loop tiling by means of cache miss equations and genetic algorithms,” 2001, URL citeseer.csail.mit.edu/abella01nearoptimal.html. Google Scholar
12. D. Burger, T. M. Austin, and S. Bennett, Tech. Rep. CS-TR-1996-1308 (1996), URL citeseer.ist.psu.edu/burger96evaluating.html. Google Scholar
13. T. S. Karkhanis and J. E. Smith, in ISCA ’04: Proceedings of the 31st Annual International Symposium on Computer Architecture (IEEE Computer Society, Washington, DC, USA, 2004), p. 338, ISBN 0-7695-2143-6. Google Scholar
14. D. Burger, T. M. Austin, and S. W. Keckler, Perform. Eval. Rev 31, 4 (2004). Google ScholarCrossref
15. P. Koiran, M. Cosnard, and M. H. Garzon, Theor. Comput. Sci. 132, 113 (1994). Google ScholarCrossref
16. M. Cook, Complex Syst. 15, 1 (2004). Google Scholar
17. J. Kilian and H. Siegelmann, in Proceedings of the 6th Annual ACM Conference on Computational Learning Theory, Santa Cruz, CA, USA, 1993. Google Scholar
18. H. Siegelmann and E. Sontag, Appl. Math. Lett. 4, 77 (1991). Google ScholarCrossref
19. M. Branicky, Theor. Comput. Sci. 138, 67 (1995). Google ScholarCrossref
20. S. Omohundro, Physica D 10, 128 (1984). Google ScholarCrossref
21. J. Ŝíma and P. Orponen, Neural Comput. 15, 693 (2003). Google ScholarCrossref
22. R. Hegger, H. Kantz, and T. Schreiber, Chaos https://doi.org/10.1063/1.166424 9, 413 (1999). Google ScholarScitation
23. T. Schreiber, Phys. Rep. https://doi.org/10.1016/S0370-1573(98)00021-0 308, 2 (1999). Google ScholarCrossref
24. Y. Yamamoto, Modern Techniques in Neuroscience Research (Springer-Verlag, Berlin, 1999), pp. 669–687. Google ScholarCrossref
25. C. -K. Peng, S. Havlin, H. Stanley, and A. Goldberger, Chaos https://doi.org/10.1063/1.166141 5, 82 (1995). Google ScholarScitation
26. G. Rangarajan and M. Ding, Phys. Rev. E https://doi.org/10.1103/PhysRevE.61.4991 61, 4991 (2000). Google ScholarCrossref
27. F. Takens, in Dynamical Systems and Turbulence, Vol. 898 of Lecture Notes in Math., edited by D. Rand and L. -S. Young (Springer-Verlag, Berlin, 1981), pp. 366–381. Google ScholarCrossref
28. N. Packard, J. P. Crutchfield, J. D. Farmer, and R. S. Shaw, Phys. Rev. Lett. https://doi.org/10.1103/PhysRevLett.45.712 45, 712 (1980). Google ScholarCrossref
29. M. Casdagli, S. Eubank, J. D. Farmer, and J. Gibson, Physica D https://doi.org/10.1016/0167-2789(91)90222-U 51, 52 (1991). Google ScholarCrossref
30. M. Rosenstein, J. Collins, and C. D. Luca, Physica D https://doi.org/10.1016/0167-2789(93)90009-P 65, 117 (1993). Google ScholarCrossref
31. H. Abarbanel, Analysis of Observed Chaotic Data (Springer-Verlag, New York, 1996). Google ScholarCrossref
32. H. Kantz and T. Schreiber, Nonlinear Time Series Analysis (Cambridge University Press, Cambridge, 1996). Google Scholar
33. M. Kennel, R. Brown, and H. Abarbanel, Phys. Rev. A https://doi.org/10.1103/PhysRevA.45.3403 45, 3403 (1992). Google ScholarCrossref
34. J. -P. Eckmann, S. O. Kamphorst, and D. Ruelle, Europhys. Lett. 5, 973 (1987). Google ScholarCrossref
35. J. Zbilut, N. Thomasson, and C. Webber, Med. Eng. Phys. https://doi.org/10.1016/S1350-4533(01)00112-6 24, 53 (2002). Google ScholarCrossref
36. M. Thiel, M. Romano, J. Kurths, and P. Read, Chaos https://doi.org/10.1063/1.1667633 14, 234 (2004). Google ScholarScitation
37. J. Theiler, J. Opt. Soc. Am. A 7, 1055 (1990). Google ScholarCrossref
38. H. Kantz, Phys. Lett. A 185, 177 (1994). Google ScholarCrossref
39. T. Schreiber and A. Schmitz, Physica D https://doi.org/10.1016/S0167-2789(00)00043-9 142, 346 (2000). Google ScholarCrossref
40. S. V. Buldyrev, A. L. Goldberger, S. Havlin, R. N. Mantegna, M. E. Matsa, C. -K. Peng, M. Simons, and H. E. Stanley, Phys. Rev. E https://doi.org/10.1103/PhysRevE.51.5084 51, 5084 (1995). Google ScholarCrossref
41. C. Heneghan and G. McDarby, Phys. Rev. E https://doi.org/10.1103/PhysRevE.62.6103 62, 6103 (2000). Google ScholarCrossref
42. E. Bradley and R. Mantilla, Chaos https://doi.org/10.1063/1.1488255 12, 596 (2002). Google ScholarScitation
43. H. D. I. Abarbanel, R. Brown, and M. B. Kennel, J. Nonlinear Sci. 1, 175 (1991). Google ScholarCrossref
44. A. Wolf, J. Swift, H. Swinney, and J. Vastano, Physica D https://doi.org/10.1016/0167-2789(85)90011-9 16, 285 (1985). Google ScholarCrossref
45. S. Wolfram, A New Kind of Science (Wolfram Media, 2002), URL http://www.wolframscience.com. Google Scholar
46. A. Seznec and N. Sendrier, ACM Trans. Model. Comput. Simul. 13, 334 (2003). Google ScholarCrossref
47. R. E. Wunderlich, T. F. Wenisch, B. Falsafi, and J. C. Hoe, in ISCA ’03: Proceedings of the 30th Annual International Symposium on Computer Architecture (ACM Press, New York, NY, 2003), pp. 84–97, ISBN 0-7695-1945-8. Google Scholar
48. T. Sherwood, E. Perelman, G. Hamerly, and B. Calder, SIGOPS Oper. Syst. Rev. 36, 45 (2002). Google ScholarCrossref
49. E. Perelman, G. Hamerly, M. V. Biesbrouck, T. Sherwood, and B. Calder, in ACM SIGMETRICS the International Conference on Measurement and Modeling of Computer Systems, 2003. Google Scholar

Chaos in computer performance

REFERENCES
Section:

Article Metrics

Altmetric

Resources

General Information

52

REFERENCESSection:ChooseTop of pageABSTRACTI.INTRODUCTIONII.PROGRAM TRACESIII.TIME SERIES METHODSIV.RESULTSV.DISCUSSIONREFERENCES <<CITING ARTICLES

Article Metrics

Altmetric

REFERENCES
Section: