NOTE: This code is not maintained anymore.

DARE

To achieve the goal of supporting flexible, extensible and interoperable usage of heterogeneous performance distributed computing (HPDC) resources, we have developed the Distributed Application Runtime Environment (DARE) framework. DARE is a standards-based, abstraction-driven middleware layer that provides these capabilities.

At the core of DARE lies SAGA and SAGA-BigJob. SAGA-BigJob is a flexible general purpose pilot-job implementation, which has been shown to be a powerful abstraction for resource management. SAGA provides the interoperability layer. Combined, SAGA and SAGA-BigJob are used to support the resource management capabilities of many tools.

Furthermore, with a suitable Web development framework, DARE supports the development of a lightweight but extensible, science gateway capable of using a range of distributed resources. Currently, DARE uses DJANGO.

DARE is an example of a Platform Independent Library, with specific support for a range of commonly occuring "patterns" (namely stand-alone applications (Type I), pipelines of tasks (Type II) and execution patterns such as bag-of-tasks, coupled-ensembles etc (Type III)

The effectiveness of the DARE framework lies in providing a simple, scalable and sustainable
approach to developing and supporting a wide range of patterns of execution independent of the infrastructure, problem instance configurations and size.

Web Page & Mailing List

Web page: https://github.com/saga-project/DARE/wiki

Mailing list: Use [email protected]

DEMO PAGE

http://gw68.quarry.iu.teragrid.org

INSTALLATION

1. Check out the code

    $ mkdir workspace 
    $ cd workspace
    $ git clone https://github.com/saga-project/DARE/
   

2. Create virtualenv

    $ virtualenv /tmp/envdare
    $ source /tmp/envdare/bin/activate
   

3. INSTALL DARE-BIGJOB Env.

    $ cd DARE/DARE-BIGJOB
    $ pip install -r requirements.txt
   

USAGE

1. create/sync/update Database schema

    $ python manage.py syncdb
    $ python manage.py migrate
   

2. Running the django web server

    $ python manage.py runserver