Distributed and parallel Python

 ppft is a friendly fork of Parallel Python. It provides:
 .
  * parallel execution of Python code on SMP and clusters
  * easy-to-understand job-based parallelization
  * automatic detection of the number of effective processors
  * dynamic processor allocation (at runtime)
  * low overhead for jobs with the same function (through transparent caching)
  * dynamic load balancing (jobs are distributed at runtime)
  * fault-tolerance (if a node fails, tasks are rescheduled on the others)
  * auto-discovery of computational resources
  * dynamic allocation of computational resources
  * SHA based authentication for network connections
  * enhanced serialization, using dill.source
 .
 ppft is part of pathos, a Python framework for heterogeneous computing.