An Analysis of Dynamic Branch Prediction Schemes on System Workloads

Recent studies of dynamic branch prediction schemes rely almost exclusively on user-only simulations to evaluate performance. We find that an evaluation of these schemes with user and kernel references often leads to different conclusions. By analyzing our own Atom-generated system traces and the system traces from the Instruction Benchmark Suite, we quantify the effects of kernel and user interactions on branch prediction accuracy. We find that user-only traces yield accurate prediction results only when the kernel accounts for less than 5% of the total executed instructions. Schemes that appear to predict well under user-only traces are not always the most effective on full-system traces: the recently-proposed two-level adaptive schemes can suffer from higher aliasing than the original per-branch 2-bit counter scheme. We also find that flushing the branch history state at fixed intervals does not accurately model the true effects of user/kernel interaction.

[1]  Chris H. Perleberg,et al.  Branch Target Buffer Design and Optimization , 1993, IEEE Trans. Computers.

[2]  James E. Smith,et al.  A study of branch prediction strategies , 1981, ISCA '98.

[3]  Yale N. Patt,et al.  A two-level approach to making class predictions , 2003, 36th Annual Hawaii International Conference on System Sciences, 2003. Proceedings of the.

[4]  Michael D. Smith,et al.  Improving the accuracy of static branch prediction using branch correlation , 1994, ASPLOS VI.

[5]  William J. Bolosky,et al.  Mach: A New Kernel Foundation for UNIX Development , 1986, USENIX Summer.

[6]  Joseph T. Rahmeh,et al.  Improving the accuracy of dynamic branch prediction using branch correlation , 1992, ASPLOS V.

[7]  Trevor N. Mudge,et al.  Instruction fetching: Coping with code bloat , 1995, Proceedings 22nd Annual International Symposium on Computer Architecture.

[8]  Alan Jay Smith,et al.  Branch Prediction Strategies and Branch Target Buffer Design , 1995, Computer.

[9]  J. Bradley Chen,et al.  Kernel instrumentation tools and techniques , 1995 .

[10]  Brian N. Bershad,et al.  The impact of operating system structure on memory system performance , 1994, SOSP '93.

[11]  Andreas Krall,et al.  Improving semi-static branch prediction by code replication , 1994, PLDI '94.

[12]  J. Bradley Chen,et al.  Software methods for system address tracing , 1993, Proceedings of IEEE 4th Workshop on Workstation Operating Systems. WWOS-III.

[13]  Yale N. Patt,et al.  A Comparison Of Dynamic Branch Predictors That Use Two Levels Of Branch History , 1993, Proceedings of the 20th Annual International Symposium on Computer Architecture.

[14]  Ravi Nair,et al.  Dynamic path-based branch correlation , 1995, MICRO 28.

[15]  Trevor N. Mudge,et al.  The role of adaptivity in two-level adaptive branch prediction , 1995, Proceedings of the 28th Annual International Symposium on Microarchitecture.

[16]  Michael D. Smith,et al.  A comparative analysis of schemes for correlated branch prediction , 1995, Proceedings 22nd Annual International Symposium on Computer Architecture.

[17]  Amitabh Srivastava,et al.  Analysis Tools , 2019, Public Transportation Systems.

[18]  S. McFarling.  Combining Branch Predictors , 1993 .