Toward Optimizing Enzo, an AMR Cosmology Application

  • James Bordner
Conference paper
Part of the Lecture Notes in Computational Science and Engineering book series (LNCSE, volume 41)


Enzo is a parallel hybrid SAMR / N-body code designed to simulate cosmological structure formation. This paper describes our approach to gathering and visualizing performance information from Enzo, which will be used to direct our subsequent modeling and optimization effort. Understanding the performance of AMR applications on distributed memory architectures is challenging, owing in part to the dynamic multilevel data structures and variety of communication patterns involved. To facilitate the task of measuring, modeling, and optimizing Enzo’s performance, we are developing the Enzo Performance Monitoring System (EPMS). We review some existing performance tools, describe the EPMS, and show some preliminary performance data obtained using the EPMS.


Load Balance Loaded Processor User Counter Storage Access High Performance Computing Application 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    G. Bryan and M.L. Norman, A Hybrid AMR Application for Cosmology and Astrophysics, in Workshop on Structured Adaptive Mesh Refinement Grid Methods, N. Chrisochoides (ed), March 1997Google Scholar
  2. 2.
    Berger and P. Collela, Local Adaptive mesh refinement for Shock Hydrodynamics, J. Comp. Phys., 82:64–84, 1989CrossRefGoogle Scholar
  3. 3.
    Daniel A. Reed and Ruth A. Aydt, Tools for Performance Tuning and Debugging, in Sourcebook of Parallel Computing, J. Dongarra (ed), I. Foster, G. Fox (ed), W. Gropp, K. Kennedy, L. Torczon and A. White (ed), Morgan Kaufman Publishers, 2003.Google Scholar
  4. 4.
    O. Zaki, E. Lusk, W. Gropp, and D. Swider, Toward Scalable Performance Visualization with Jumpshot, Int’l J. of High Performance Computing Applications, 1999.Google Scholar
  5. 5.
    Luiz DeRose and Daniel A. Reed, SvPablo: A Multi-Language Architecture-Independent Performance Analysis System, Proceedings of the International Conference on Parallel Processing (ICPP’99), Fukushima, Japan, September 1999.Google Scholar
  6. 6.
    Z. Lan, V. Taylor, and G. Bryan, A Novel Dynamic Load Balancing Scheme for Parallel Systems, Journal of Parallel and Distributed Computing, Vol 62/12, pp.1763–1781, 2002.CrossRefGoogle Scholar
  7. 7.
    Z. Lan, V. Taylor, and G. Bryan, Dynamic Load Balancing of SAMR applications on Distributed Systems, Journal of Scientic Programming, Vol 10(4), pp. 319–328, 2002.Google Scholar
  8. 8.
    Lorie M. Liebrock, Using Problem Topology in Parallelization, Rice University Technical Report CRPC-TR94477-S, September 1994.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2005

Authors and Affiliations

  • James Bordner
    • 1
  1. 1.University of CaliforniaSan DiegoUSA

Personalised recommendations