jug
Jug is a task-based parallelism framework. Jug allows you to write code that is broken up into tasks and run different tasks on different processors. It uses the filesystem to communicate between processes and works correctly over NFS, so you can coordinate processes on different machines. Jug is a pure Python implementation and should work on any platform that can run Python.
| Tags | Python parallel programming mapreduce |
|---|---|
| Licenses | MIT |
| Operating Systems | Unix OS Independent |
| Implementation | Python |
Recent releases
- 06 Dec 2011 03:28
Release Notes: The most important change in this release is re-running of the jugfile on barrier(). A few convenience functions were added and bugs were fixed.
- 06 Jul 2011 07:22
Release Notes: Several bugs were fixed, including for compatibility with new versions of external packages (ipython and redis). Input/Output of very large files is now improved so that it potentially uses much less intermediate memory. Unit tests were added to the distribution. A non-compatible change is that numpy arrays are now hashed differently. However, the new method is much, much better (faster and uses less memory); so it was worth breaking compatibility.
- 28 Mar 2011 21:16
Release Notes: This release finishes work on Tasklets. There are many bugfixes since the Beta release (in the cleanup subcommand, status with --cache, and CompoundTask).
- 10 Mar 2011 21:56
Release Notes: This release introduces Tasklets, which are lightweight tasks that do not save their results to disk (but rather compute them whenever needed).
- 17 Jan 2011 04:37
Release Notes: This release fixes a case where ~/.jug/configrc did not exist. It will print the host name to the lock file on file_store. Refactored implementation of options. Unloading tasks that have not run has been fixed. mapreduce has been fixed regarding empty input.
Heartbeat
- Popularity Score
- 98.21
- Vitality Score
- 8.43
- Subscriptions
- 14
