Accurate Cache and TLB Characterization Using Hardware Counters

  • Jack Dongarra
  • Shirley Moore
  • Philip Mucci
  • Keith Seymour
  • Haihang You
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3038)


We have developed a set of microbenchmarks for accurately determining the structural characteristics of data cache memories and TLBs. These characteristics include cache size, cache line size, cache associativity, memory page size, number of data TLB entries, and data TLB associativity. Unlike previous microbenchmarks that used time-based measurements, our microbenchmarks use hardware event counts to more accurately and quickly determine these characteristics while requiring fewer limiting assumptions.


Cache Size Array Size Data Cache Page Size Cache Associativity 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. 1.
    Intel Itanium 2 Processor Reference Manual (April 2003),
  2. 2.
    Bilmes, J., Asanovic, K., Chin, C.-W., Demmel, J.: Optimizing matrix multiply using PhiPAC: a portable high-performance ANSI C coding methodology. In: Proc. International Conference on Supercomputing, Vienna, Austria (1997)Google Scholar
  3. 3.
    Browne, S., Dongarra, J., Garner, N., Ho, G., Mucci, P.: A portable programming interface for performance evaluation on modern processors. International Journal of High Performance Computing Applications 14(3), 189–204 (2000)CrossRefGoogle Scholar
  4. 4.
    Chilimbi, T.M., Hill, M.D., Larus, J.D.: Cache-conscious structure layout. In: Proc. 1999 ACM SIGPLAN Conference on Programming Languages and Implementation (PLDI), pp. 1–12 (1999)Google Scholar
  5. 5.
    Jalby, W., Lemuet, C.: Exploring and optimizing Itanium2 cache performance for scientific computing. In: Proc. 2nd Workshop on EPIC Architectures and Compiler Technology, Istanbul, Turkey (November 2002)Google Scholar
  6. 6.
    Manegold, S., Boncz, P.: Cache-Memory and TLB Calibration Tool (2001),
  7. 7.
    Saavedra, R.H., Smith, A.J.: Measuring cache and TLB performance and their effect on benchmark runtimes. IEEE Transactions on Computers 44(10), 1223–1235 (1995)zbMATHCrossRefGoogle Scholar
  8. 8.
    Snavely, A., Carrington, L., Wolter, N., Labarta, J., Badia, R., Purkayastha, A.: A framework for performance modeling and prediction. In: Proc. SC 2002, Baltimore, MD (November 2002)Google Scholar
  9. 9.
    Sorenson, E.S., Flanagan, J.K.: Cache characterization surfaces and predicting workload miss rates. In: Proc. 4th IEEE Workshop on Workload Characterization, Austin, Texas, December 2001, pp. 129–139 (2001)Google Scholar
  10. 10.
    Thomborson, C., Yu, Y.: Measuring data cache and TLB parameters under Linux. In: Proc. 2000 Symposium on Performance Evaluation of Computer and Telecommunication Systems, pp. 383–390. Society for Computer Simulation International (July 2000)Google Scholar
  11. 11.
    Whaley, R.C., Petitet, A., Dongarra, J.: Automated empirical optimizations of software and the ATLAS project. Parallel Computing 27(1-2), 3–25 (2001)zbMATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Jack Dongarra
    • 1
  • Shirley Moore
    • 1
  • Philip Mucci
    • 1
  • Keith Seymour
    • 1
  • Haihang You
    • 1
  1. 1.Innovative Computing LaboratoryUniversity of TennesseeKnoxvilleUSA

Personalised recommendations