Understanding Asynchronous Parallel Pattern Search

  • Tamara G. Kolda
  • Virginia J. Torczon
Part of the Applied Optimization book series (APOP, volume 82)


Asynchronous parallel pattern search (APPS) is a nonlinear optimization algorithm that dynamically initiates actions in response to events, rather than cycling through a fixed set of search directions, as is the case for synchronous pattern search. This gives us a versatile concurrent strategy that allows us to effectively balance the computational load across all available processors. However, the semi-autonomous nature of the search complicates the analysis. We concentrate on elucidating the concepts and notation required to track the iterates produced by APPS across all participating processes. To do so, we consider APPS and its synchronous counterpart (PPS) applied to a simple problem. This allows us both to introduce the bookkeeping we found necessary for the analysis and to highlight some of the fundamental differences between APPS and PPS.


nonlinear optimization asynchronous parallel optimization pattern search global convergence distributed computing cluster computing 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    M. Avriel. (1976), Nonlinear Programming: Analysis and Methods,Prentice-Hall, Englewood Cliffs, New Jersey.Google Scholar
  2. [2]
    D. Bertsekas and J. Tsitsiklis. (1989), Parallel and Distributed Computation: Numerical Methods, Prentice-Hall, Englewood Cliffs, New Jersey.Google Scholar
  3. [3]
    P. D. Hough, T. G. Kolda, and V. J. Torczon (2001), “Asynchronous parallel pattern search for nonlinear optimization,” SIAM J. Scientific Computing, 23, pp. 134–156.MathSciNetzbMATHCrossRefGoogle Scholar
  4. [4]
    T. G. Kolda and V. J. Torczon (2001) “On the convergence of asynchronous parallel pattern search,” Tech. Rep. SAND2001–8696, Sandia National Laboratories, Livermore, California.Google Scholar
  5. [5]
    D. Levine (1995) “Users guide to the PGAPack parallel genetic algorithm library,” Tech. Rep. ANL-95/18, Argonne National Laboratory, Argonne, Illinois.Google Scholar
  6. [6]
    V. Torczon (1992), “PDS: Direct search methods for unconstrained optimization on either sequential or parallel machines,” Tech. Rep. 92–09, Rice University, Department of Computational and Applied Mathematics, Houston, Texas.Google Scholar
  7. [7]
    V. Torczon (1997), “On the convergence of pattern search algorithms,” SIAM J. Optim.,7, pp. 1–25.MathSciNetzbMATHCrossRefGoogle Scholar

Copyright information

© Kluwer Academic Publishers B.V. 2003

Authors and Affiliations

  • Tamara G. Kolda
    • 1
  • Virginia J. Torczon
    • 2
  1. 1.Computational Sciences and Mathematics Research DepartmentSandia National LaboratoriesLivermoreUSA
  2. 2.Department of Computer ScienceCollege of William & MaryWilliamsburgUSA

Personalised recommendations