https://emews.org/emews-tutorial
The EMEWS tutorial presents an overview of EMEWS, the Swift/T parallel scripting language, EMEWS templates, and a number of use-cases, starting with a simple agent-based model parameter sweep, and ending with a complex adaptive parameter space exploration workflow coordinating ensembles of distributed (MPI) simulations. The use-cases are available for interested parties to download and run on their own. While the example models in the tutorial utilize agent-based models, EMEWS can be applied to any computational modeling method requiring heuristic model exploration.
https://github.com/emews/emews-project-creator
EMEWS Creator is a Python application for creating workflow projects for EMEWS from the command line. The project consists of the canonical EMEWS directory layout and various files that can be customized by the user for their particular use case. Further information on the EMEWS Creator can be found in the EMEWS Creator section of the EMEWS tutorial.
EQ/R is an R-based Swift/T resident task extension that allows Swift/T workflows to communicate with a persistent embedded R interpreter on a worker process via blocking queues. Using EQ/R, an R-based ME algorithm can be used to control and define an ensemble of model runs. More information can be found in the EQ/R section of the EMEWS tutorial.
https://github.com/emews/EQ-Py
EQ/Py is a Python-based Swift/T resident task extension that allows Swift/T workflows to communicate with a persistent embedded Python interpreter on a worker process via blocking queues. Using EQ/Py, a Python-based ME algorithm can be used to control and define an ensemble of model runs. More information can be found in the EQ/Py section of the EMEWS tutorial.
https://github.com/emews/EQ-SQL
EQ-SQL is our newest decoupled architecture and task API for distributing workflows on heterogeneous computing resources. It provides more flexibility and robustness for running longer and heterogeneous workflows. More information can be found in the EQ-SQL section of the EMEWS tutorial.
https://github.com/emews/emews_tutorial_BO
These are worked examples of optimizing a simple simulation model (a Zombies demonstration model, distributed with Repast4Py) using EQ-SQL. Both Python and R Bayesian optimization code are demonstrated, with both a local and remote deployment.
The EMEWS Model Exploration Library Archive (MELA) is a collection of model exploration (ME) modules, which can be directly incorporated into EMEWS workflows. The ME modules are written in R or Python (used with EQ/R and EQ/Py, respectively) and come with descriptions of the underlying ME algorithm, the communication protocol and the ability to run standalone (i.e., in their original language) ME tests.