Speakers
Report on the experience (or the proposed activity). It would be very important to mention key services which are essential for the success of your activity on the EGEE infrastructure.
We demonstrate JP usage in two cases. In the first one, management of
large parametric studies in computational chemistry (molecular docking)
JP, together with a thin graphical front-end, was used to build the job
catalogue from scratch. It allowed the researchers to easily manipulate
computational jobs (input modification, jobs resubmission), to search and
selected desired (finished/non-finished, aborted) jobs and finally
utilize specific plugins for results presentation (e.g. visualization).
In the second case, we augmented production jobs of the Atlas experiment
to interact with JP, yielding functionality similar to Atlas ProdDB but
with emphasis in job history. We routed part of the Atlas production
traffic to JP (approx. 1100 jobs/day) as well as performed stress tests
on snapshot of these jobs in order to demonstrate JP readiness for
production deployment.
Describe the scientific/technical community and the scientific/technical activity using (planning to use) the EGEE infrastructure. A high-level description is needed (neither a detailed specialist report nor a list of references).
Here we demonstrate a generic gLite middleware service - Job Provenance -
providing a backbone for custom application solutions requiring job cataloque
capabilities. The target communities vary from small research groups
trying to set up their own solution fulfilling their specific needs up to
potential usage in well established communities like the high energy particle
experiments.
Describe the added value of the Grid for the scientific/technical activity you (plan to) do on the Grid. This should include the scale of the activity and of the potential user community and the relevance for other scientific or business applications
gLite Job Provenance (JP) is a generic job catalogue service keeping
long-term track of execution of Grid jobs. It provides a sophisticated
machinery to support application and user annotations of the Grid
computational jobs. Furthermore it provides data mining over the raw data and
annotations. Being a standard part of gLite middleware stack it
offers continuous and guaranteed service to store all the primary
information. On top of storage facilities, Job Provenance Index Servers
allow for efficiently looking for expected and unexpected patterns within
the stored information through user queries. While JP can be used
directly by all gLite middleware users, specialized job catalogues can be
built with moderate effort compared to custom solutions (custom job
catalogue development) taking considerable effort.
