Speakers
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
DIOGENES (DIstributed Optimal GENEtic algorithm for grid application
Scheduling) provides a solution for Grid meta-scheduling, aiming
to achieve a
distributed, fault-tolerant, scalable and efficient method for
optimizing task
assignment in Grid. The DIOGENES scheduler uses a combination of
genetic
algorithms and lookup services for obtaining a scalable and
highly reliable
optimization tool. Gridification topics that are covered are:
application
monitoring, running job, data management and access, workflows.
The social
impact is more efficient use of SEE-GRID VO’s resources and the
scientific impact
represent new optimization method for Grid scheduling using Genetic
Algorithms The current research is focused on multi-criteria
optimization of Grid
scheduling, complex task dependencies, backup and recovery from
service
failures (re-scheduling). We also provide a method for efficient
integration with
the Globus Toolkit, based on Open Grid Service Architecture.
With a forward look to future evolution, discuss the issues you have encountered (or that you expect) in using the EGEE infrastructure. Wherever possible, point out the experience limitations (both in terms of existing services or missing functionality)
EGEE includes the services requested by DIOGENES. For the moment
we don't
foresee difficulties in using it on this infrastructure. We’ll
use DIOGENES in
projects like GridMOSI – VO in Grid technology for high
performance MOdeling,
SImulation, and optimization, and MedioGRID – Parallel and
distributed
processing of geographic and environmental data. We will pre-test
new
strategies developed in DIOGENES for applications built in these
and other
projects and eventually run on EGEE.
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.
DIOGENES doesn't control the clusters / resources directly. It is
more natural to
consider it closer to Grid applications. It is responsible for
the management of
jobs, such as allocating resources needed for any specific job,
managing the
tasks for parallel execution, managing of data, and correlation
of events. To
fulfill its functions, DIOGENES needs information coming from
information and
monitoring services available in the platform, in this case in
EGEE. The results of
testing DIOGENES strategies in a Romanian cluster of SEE-GRID
demonstrate
very good behaviour in comparison with other scheduling approaches.
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).
The work is developed as a Grid application/service in the
See-Grid project, by a
team in the National Center for Information Technology (NCIT) of the
University “Politehnica” of Bucharest. NCIT is member of the
RoGrid consortium,
which participates in See-Grid and in EGEE-2. NCIT is involved in
projects that
are developed in cooperation with CERN and Caltech (RoDiCA),
develops
national Grid research projects GridMOSI, MedioGRID, and
contributes to See-
Grid and EGEE projects.
