| Tuesday, February 26 |
| 3:30 pm - 5:00 pm | |
| |
Scaling DRMAA codes to the Grid: A Demonstration on EGEE, TeraGrid and OSG.
(90 mins)
Eduardo Huedo and Tino Vázquez (http://www.dsa-research.org)
View Participants
The aim of the presentation is to describe the DRMAA language bindings provided by GridWay (C, JAVA, Perl, Python and Ruby) and its support for interoperability, being able to interface with various infrastructures running different middlewares. GridWay DRMAA implementation allows applications intended for clusters to scale to grid environments. As use case, we will demonstrate a widely used Bioinformatics application, CD-HIT, running on the EGEE, TeraGrid and OSG Infrastructures.
A. The DRMAA standard
DRMAA Specification 1.0 is an OGF Grid Recommendation, acting as the normative base for all DRMAA language bindings. DRMAA provides application developers and distributed resource management builders with a programming model that enables the development of distributed applications tightly coupled to an underlying DRMS. DRMAA is implemented and therefore usable with the following commercial or freely available DRM systems: Condor, LSF, GridWay, Grid Engine and PBS / Torque.
B. The GridWay metascheduler
GridWay is a widely-used metascheduling technology that performs job execution management and resource brokering, allowing unattended, reliable, and efficient execution of jobs, job arrays, and workflows on heterogeneous and dynamic Globus Grids. The GridWay metascheduler is a Globus product, released under Apache license v2.0, that implements different OGF standards, such as DRMAA or JSDL. The modular design of GridWay allows its integration with the resource management, file management and information services available in a given infrastructure. For example, GridWay is fully functional on EGEE, TeraGrid and OSG infrastructures
C. The CD-HIT application
CD-HIT application performs protein clustering, which consists in removing redundant sequences from a protein database in order to generate a database of only the representatives. Protein clustering can be applied in many activities such as protein family classification, domain analysis, organization of large protein databases or improving database search performance. CD-HIT has been ported to the Grid using DRMAA (Distributed Resource Management Application API) in order to assure its compatibility with resource managers that implement the standard.
Agenda:
1. Introduction to DRMAA
2. DRMAA bindings provided by GridWay
3. GridWay approach to interoperability
4. Demonstration of CD-HIT running on EGEE, OSG and TeraGrid resources
Location: Molly Pitcher
|
| |
| |
|
| |
| | Slides: Scaling DRMAAA codes to the Grid |