pomsets has many features to support the execution of your computational workflows in the cloud:

• Parallel computing: pomsets enables you to run your workflows in single threaded mode and in parallel if you have additional processors available.
• Data flow: pomsets allows you to define data dependencies in your workflows
• Workflow reusability: pomsets enables you to reuse the workflows. You can run them unmodified on the same dataset or specify a different dataset for each run. You can also share the pomsets with your colleagues.
• Compute cloud agnosticism: pomsets is agnostic with respect to your compute cloud deployment. pomsets currently supports your workflow execution on both Amazon's EC2 and Eucalyptus clouds. More compute cloud APIs are forthcoming
• Execute environment agnosticism: pomsets is agnostic with respect to the environment for your workflow. Direct local and remote command line execution are supported out of the box. So are runtime Python evaluations and executing Hadoop jobs. Support for various grid engines and distributed task queues are forthcoming.
• MapReduce: pomsets supports the MapReduce paradigm in an agnostic manner. You can use the internal parameter sweep mechanism or connect to your Hadoop installation.
• Flow control: pomsets supports flow control operators such as parameter sweeps, loops, and branches.
• Intuitive GUI: The pomsets GUI has been designed to be as easy to use as possible.
• Simple API: pomsets has a simple API that enables your applications to programmatically create and execute workflows.

Feature comparison

Also available are feature comparisons against other workflow management systems.