About
Parampl is a simple Python tool for running AMPL programs in parallel, and across multiple machines when needed.
From within your AMPL code, you can dispatch subproblems asynchronously to separate threads, synchronize execution, and coordinate results back into the program flow. That lets you use multiple CPU cores while solving, or spread work across a cluster.
For modellers working with large or structured problems, Parampl makes it possible to decompose a task into subproblems and express solution algorithms as a series of coordinated subtasks.
Related papers
-
Parampl: A Simple Tool for Parallel and Distributed Execution of AMPL Programs
IEEE Access, Volume 6, pp. 49282-49291, IEEE, September 2018. DOI: 10.1109/ACCESS.2018.2868222
If you use Parampl in your research, please cite this paper. -
Parampl: A simple approach for parallel execution of AMPL programs
10th International Conference on Parallel Processing and Applied Mathematics (PPAM 2013), Part II, LNCS 8385, pp. 86-94, 2014.
Available at Springer Link . DOI: 10.1007/978-3-642-55195-6_8
Downloads
To start working with Parampl, visit the project repository on GitHub for documentation, examples, and source code.