Skip to main content

JIP pipeline library

Project description

JIP Pipeline system

The JIP pipeline system is a python library and a set of command line utilities that allows you to create batch-process based computational pipeline that can be submitted and managed on a compute cluster or on your local machine.

Installation

The JIP documentation contains more detailed instructions on installation and, more importantly, configuration of the system and your compute infrastructure. Here is the very quick guide:

$> pip install pyjip

or

$> python setup.py install

If you want to install from pypi or the git repository. Afterwards, you have to create configuration file $HOME/.jip/jip.json and add the following content:

{
    "cluster": "<class_name>"
}

Replace <class_name> with the name of the class that implements support for your compute cluster. JIP ships with the following implementations:

  • jip.cluster.SGE for Sun Grid Engine

  • jip.cluster.Slurm for Slurm

  • jip.cluster.PBS for PBS and Torque

  • jip.cluster.LSF for Platform LSF and Open Lava

  • jip.grids.JIP for JIPs’ internal scheduler. If you use this, make sure to start the JIP server on the same machine using the jip server command. Please note also that you have to install pyzmq if you want to run the JIP server.

Documentation

Documentation on installation and usage can be found at http://pyjip.readthedocs.org/en/latest/.

Source Code

The JIP source code can be found on GitHub.

Bugs and feature requests

Please feel free to use the issue tracker to file bug reports and feature requests.

Changelog

0.4:
  • Create a pipe command to write quick pipeline directly from the command line [Issue 22]

  • Inconsistency between job().bash() and bash() usage in pipeline [Issue 20]

  • Add links to job and pipeline iteration to delete functions in jip.db and jip.jobs [Issue 19]

  • Enable parsing of memory assignments in profiles and add support for G M and K suffixes [Issue 18]

  • “jip logs” called without argument returns an error instead of “usage” [Issue 16]

  • Tool cleanup fails if the output points to a directory [Issue 10]

  • Option parsing fails when multiple list options are specified and one is optional and not set [Issue 8]

  • jip restart jobs fails as the job detects itself and refuses to submit again [Issue 7]

  • jip jobs with selected output duplicates columns [Issue 5]

0.3 :
  • Added auto-naming support for anonymous jobs that are assigned to variables. The variable name is used as default job name

  • Lots of changes to the internals

0.2 :
  • Added explicit UTF-8 encoding for the dependency tree

0.1 : Initial release

Licences

JIP is licensed under the BSD license.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

pyjip-0.4.tar.gz (155.7 kB view details)

Uploaded Source

File details

Details for the file pyjip-0.4.tar.gz.

File metadata

  • Download URL: pyjip-0.4.tar.gz
  • Upload date:
  • Size: 155.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No

File hashes

Hashes for pyjip-0.4.tar.gz
Algorithm Hash digest
SHA256 5e177f6092a3b08e087d66456e2782f36501748b15aa4105d34d8db4d292a64b
MD5 a32a93ba5c6f09390541aa736e0ecbef
BLAKE2b-256 22f4edcb4c62c463a36862bed68cb07d17b3ba6fa47332f7516c8ef79dbbeaba

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page