[SAGA-RG] SAGA-Python 0.9.8 Released (Including support for Amazon EC2 and PBS Job Callbacks)

[Ole Weidner]

[PLEASE DISTRIBUTE]

Dear SAGA-Python Community,

you have probably already noticed that we have pushed out multiple version of saga-python to PyPi in the last three weeks in rapid succession: 0.9.4, 0.9.5, 0.9.6, 0.9.7, and 0.9.8.

While more than half of them are incremental bugfix-releases, there are also a great number of new features part of the latest saga-python release. Here's a quick overview:

1. Resource Package
-------------------

SAGA-Python 0.9.8 is the first SAGA implementation to provide the SAGA Resource Package API. The resource package abstracts the task of resource provisioning, allocation and management and maps well to distributed computing concepts like 'cloud computing'/IaaS and 'advanced reservation'. The resource package is still work in progress, but some rudimentary documentation along with an example can be found here: http://saga-project.github.io/saga-python/doc/library/resource 

2. 'Libcloud' / Amazon EC2 Adaptor
----------------------------------

The first adaptor that implements the new resource package is based on Apache 'libcloud' (https://libcloud.apache.org/). It currently supports and has been tested with Amazon EC2 cloud services. It allows to instantiate new EC2 VMs from any of the images / types available on EC and to list and shutdown existing ones. The libcloud adaptor ties in nicely with the SSH job adaptor: it allows an application to allocate EC2 resources and then transparently submit 'jobs' to these resources through the same API. A complete example can be found here: http://saga-project.github.io/saga-python/doc/adaptors/saga.adaptor.ec2_resource.html 

3. Callback support for PBS
---------------------------

As the first saga-python adaptor, the PBS adaptor now supports job state callbacks as defined in the SAGA GFD.90 specification. This has been a long-standing item on the saga wish list as it can potentially simplify application development significantly while providing better performance. An example how to use callbacks with the PBS adaptor can be found here: https://github.com/saga-project/saga-python/blob/master/examples/jobs/pbsjob.py. After we get some feedback about the current implementation, we plan to add callback support to the remaining job adaptors over the next couple of releases. 

We have also updated the saga-python website (http://saga-project.github.io/saga-python/), the API documentation (http://saga-project.github.io/saga-python/doc) and some of the examples and tutorial material. 

The full change log can be found here: https://github.com/saga-project/SAGA-Python/blob/master/CHANGES.md. 

A big 'thank you' to everyone who contributed to the - at times painful - development and testing efforts over the past couple of weeks. I believe that this release really marks the point at which saga-python is not only a clean re-implementation of the previous 'saga-cpp' and 'saga-bliss' packages, but has also surpassed them in terms of functionality, stability and features! 

Please let us know if you encounter any problems with this release or if you have suggestions and idea for improvement for future versions of saga-ython.

Best regards,
Ole