Dynamo: a transparent dynamic optimization system

ABSTRACT

We describe the design and implementation of Dynamo, a software dynamic optimization system that is capable of transparently improving the performance of a native instruction stream as it executes on the processor. The input native instruction stream to Dynamo can be dynamically generated (by a JIT for example), or it can come from the execution of a statically compiled native binary. This paper evaluates the Dynamo system in the latter, more challenging situation, in order to emphasize the limits, rather than the potential, of the system. Our experiments demonstrate that even statically optimized native binaries can be accelerated Dynamo, and often by a significant degree. For example, the average performance of -O optimized SpecInt95 benchmark binaries created by the HP product C compiler is improved to a level comparable to their -O4 optimized version running without Dynamo. Dynamo achieves this by focusing its efforts on optimization opportunities that tend to manifest only at runtime, and hence opportunities that might be difficult for a static compiler to exploit. Dynamo's operation is transparent in the sense that it does not depend on any user annotations or binary instrumentation, and does not require multiple runs, or any special compiler, operating system or hardware support. The Dynamo prototype presented here is a realistic implementation running on an HP PA-8000 workstation under the HPUX 10.20 operating system.

References

1. 1.Auslander, J., Philipose, M., Chambers, C., Eggers, S.J., and Bershad, B.N. 1996. Fast, effective dynamic compilation. In Proceedings of the SIGPLAN'96 Conference on Programming Language Design and Implementation (PLDI'96).
2. 2.Bala, V., Duesterwald, E., and Banerjia, S. 1999. Transparent dynamic optimization: The design and implementation of Dynamo. Hewlett Packard Laboratories Technical Report HPL-1999-78. June 1999.
3. 3.Bala V., and Freudenberger, S. 1996. Dynamic optimization: the Dynamo project at HP Labs Cambridge (project proposal). HP Labs internal memo, Feb 1996.
4. 4.Ball, T., and Larus, J.R. 1996. Efficient path profiling. In Proceedings of the 29th Annual International Symposium on Microarchitecture (MICRO-29), Paris. 46-57.
5. 5.Bedichek, R. 1995. Talisman: fast and accurate multicomputer simulation. In Proceedings of the 1995 ACM SIGMETRICS Conference on Measurement and Modeling of Computer Systems.
6. 6.Chambers, C., and Ungar, D. 1989. Customization: optimizing compiler technology for Self, a dynamically-typed object-orientied programming language. In Proceedings of the SIGPLAN'89 Conference on Programming Language Design and Implementation. 146-160.
7. 7.Chernoff, A., Herdeg, M., Hookway, R., Reeve, C., Rubin, N., Tye, T., Yadavalli, B., and Yates, J. 1998. FX!32: a profile-directed binary translator. IEEE Micro, Vol 18, No. 2, March/April 1998.
8. 8.Cmelik, R.F., and Keppel, D. 1993. Shade: a fast instruction set simulator for execution profiling. Technical Report UWCSE-93-06-06, Dept. Computer Science and Engineering, University .of Washington.
9. 9.Consel, C., and Noel, F. 1996. A general approach for runtime specialization and its application to C. In Proceedings of the 23th Annual Symposium on Principles of Programming Languages. 145-156.
10. 10.Cramer, T., Friedman, R., Miller, T., Seberger, D., Wilson, R., and Wolczko, M. 1997. Compiling Java Just In Time. IEEE Micro, May/Jun 1997.
11. 11.Deutsch, L.P. and Schiffman A.M. 1984. Efficient implementation of the Smalltalk-80 system. In Proceedings of the 11th Annual ACM Symposium on Principles of Programming Languages. 297-302.
12. 12.Ebcioglu K., and Altman, E.R. 1997. DAISY: Dynamic compilation for 100% architectural compatibility. In Proceedings of the 24th Annual International Symposium on Computer Architecture. 26-37.
13. 13.Engler, D.R. 1996. VCODE: a retargetable, extensible, very fast dynamic code generation system. In Proceedings of the SIGPLAN'96 Conference on Programming Language Design and Implementation (PLDI' 96).
14. 14.Fisher, J., and Freudenberger, S. 1992. Predicting conditional branch directions from previous runs of a program. In Proceedings of the 5th International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS 5). Oct 1992.85-95.
15. 15.Friendly, D.H., Patel, S.J., and Part., Y.N. 1998. Putting the fill unit to work: dynamnic optimizations for trace cache microprocessors. In Proceedings of the 31st Annual Internation Symposium on Microarchitecture (MICRO-31), Dallas. 173-181.
16. 16.Grant, B., Philipose, M., Mock, M., Chambers, C., and Eggers, S.J. An evaluation of staged run-time optimizations in DyC. In Proceedings of the SIGPLAN'99 Conference on Programming Language Design and Implementation. 293- 303.
17. 17.Herold, S.A. 1998. Using complete machine simulation to understand computer system behavior. Ph.D. thesis, Dept. Computer Science, Stanford University.
18. 18.Hwu, W.W., Mahlke, S.A., Chen, W.Y., Chang, P. P., Warter, N.J., Bringmann, R.A., Ouellette, R.Q., Hank, R.E., Kiyohara, T., Haab, G.E., Holm, J.G., and Lavery, D.M. 1993.The superblock: an effective structure for VLIW and superscalar compilation. The Journal of Supercomputing 7, (Jan.). 229- 248.
19. 19.Keller, J. 1996. The 21264: a superscalar Alpha processor with out-of-order execution. Presented at the 9th Annual Microprocessor Forum, San Jose, CA.
20. 20.Kelly, E.K., Cmelik, R.F., and Wing, M.J. 1998. Memory controller for a microprocessor for detecting a failure of speculation on the physical nature of a component being addressed. U.S. Patent 5,832,205, Nov. 1998.
21. 21.Kumar, A. 1996. The HP PA-8000 RISC CPU: a high performance out-of-order processor. In Proceedings of Hot Chips VIII, Palo Alto, CA.
22. 22.Leone, M. and Dybvig, R.K. 1997. Dynamo: a staged compiler architecture for dynamic program optimization. Technical Report #490, Dept. of Computer Science, Indiana University.
23. 23.Leone, M. and Lee, P. 1996. Optimizing ML with run-time code generation. In Proceedings of the SIGPLAN'96 Conference on Programming Language Design and Implementation. 137-148.
24. 24.Marlet, R., Consel, C., and Boinot, P. Efficient incremental run-time specialization for free. In Proceedings of the SIGPLAN '99 Conference on Programming Language Design and Implementation. 281-292.
25. 25.Papworth, D. 1996. Tuning the Pentium Pro microarchitecture. IEEE Micro, (Apr.). 8-15.
26. 26.Poletta, M., Engler, D.R., and Kaashoek, M.F. 1997. tcc: a system for fast flexible, and high-level dynamic code generation. In Proceedings of the SIGPLAN '97 Conference on Programming Language Design and Implementation. 109- 121.
27. 27.Rotenberg, E., Bennett, S., and Smith, J.E. 1996. Trace cache: a low latency approach to high bandwidth instruction fetching. In Proceedings of the 29th Annual International Symposium on Microarchitecture (MICRO-29), Paris. 24-35.
28. 28.Sannella, M., Maloney, J., Freeman-Benson, B., and Borning, A. 1993. Multi-way versus one-way constraints in user interfaces: experiences with the Deltablue algorithm. Software - Practice and Experience 23, 5 (May). 529-566.
29. 29.Sites, R.L., Chernoff, A., Kirk, M.B., Marks, M.P., and Robinson, S.G. Binary Translation. Digital Technical Journal, Vol 4, No. 4, Special Issue, 1992.
30. 30.Stears, P. 1994. Emulating the x86 and DOS/Windows in RISC environments. In Proceedings of the Microprocessor Forum, San Jose, CA.
31. 31.Witchel, E. and Rosenblum R. 1996. Embra: fast and flexible machine simulation. In Proceedings of the SIGMETRICS '96 Conference on Measurement and Modeling of Computer Systems. 68-78

Index Terms

• Software and its engineering

  • Software notations and tools

    • Compilers

  • Software organization and properties

    • Contextual software domains

      • Operating systems

        • Process management