COPS: Large-Scale Optimization Problems
We are continuing the development of COPS, a
Constrained Optimization Problem Set.
The primary purpose of this collection is to provide difficult test
cases for optimization software.
Problems in the current version of the collection
come from fluid dynamics, population dynamics, optimal design,
mesh smoothing, and optimal control.
For each problem we provide a short
description of the problem, notes on the formulation of
the problem, and results of computational
experiments with general optimization
solvers. Each problem has been implemented in
AMPL. The models from COPS 2.0 are also available
in GAMS, courtesy of
GAMS Development Corporation.
The current version of the test set are documented in the report
which contains descriptions of the problems, notes
on the formulation, graphs of the solutions, and the
results of computational experiments. The article
Benchmarking Optimization Software with COPS 3.0,
Mathematics and Computer Science Division,
Argonne National Laboratory,
Technical Report ANL/MCS-273,
discusses the analysis and interpretation of benchmarking data.
The MATLAB script perf.m can
be used to create performance profiles for MATLAB. The header of the script
contains documentation on its use. Michael Friedlander has also written
performance profiling script in Python that uses gnuplot to create
encapsulated post-script files.
E. Dolan and J. Moré,
Benchmarking Optimization Software with Performance Profiles,
Mathematical Programming 91, pages 201-213, 2002
examines the importance of optimality measures
when benchmarking a set of solvers.
E. Dolan, J. Moré, and T. Munson,
Optimality measures for performance profiles,
SIAM Journal on Optimization, 16 (2006), 891 -- 909.
The COPS collection of problems can be obtained by entering your
email address below (so that you can be informed of future software
updates) and downloading the compressed tar file that contains the AMPL
model and data files (Version 3.0) or the compressed tar file of GAMS
models (Version 2.0).
COPS: Keeping optimization software honest.
Argonne National Laboratory / email@example.com