Speaker: Sudharshan Vazhkudai , Computer Science and Mathematics Division, Oak Ridge National Laboratory
Towards a Data Management Infrastructure for the Spallation Neutron Source
Abstract: The Spallation Neutron Source (SNS) at ORNL -- once operational in 2006 -- will generate several terabytes of data per year for the next few decades. The SNS infrastructure will be used by neutron scientists all over the world to run their experiments, conduct analysis on the data generated, visualize them and share their results with others in the community. Such a vast and dispersed user community brings with it diverse sets of requirements in terms of the need for different kinds of software analysis and collaborative visualization tools, a sophisticated data and metadata infrastructure, secure authentication mechanisms, user collaboration and the like. To address these issues and several others, SNS has initiated projects such as Neutron Science Software Initiative (NeSSI) and TeraGrid Extension to Neutron Science (SETENS). NeSSI is responsible for the providing an overall software solution for the neutron science community, while SETENS addresses bringing neutron source data to the NSF TeraGrid testbed.
Data Grids provide an efficient environment to handle massively distributed data by building federations of high-end storage systems, providing secure access to remote resources, enabling resource discovery, data movement, and the like. Grids provide a hierarchical framework comprising the fabric, middleware and application specific layers offering solutions for many of the aforementioned issues. However, specific science domains such as SNS demand new solutions. In this talk, I will present some of our work towards (1) building Grid middleware for efficient data access, data transfer predictions and delivery (2) revisiting the Grid storage fabric to accommodate the growing data demands of SNS by incorporating commodity workstation storage and (3) building higher-level, SNS-application specific portals.
Short Bio: Sudharshan Vazhkudai is a Scientist in the Network and Cluster Computing group of the Computer Science and Mathematics Division at the US Department of Energy's Oak Ridge National Laboratory (ORNL). His current work is on massively distributed Terascale systems. He received a Ph.D. and Masters in Computer Science from the University of Mississippi in 2003 and 1998 respectively. His doctoral work (at Argonne National Laboratory where he was on a Givens Fellowship and a Dissertation Fellowship) addressed data access issues in the Globus Data Grid environment. Prior to that, he was working on transforming Linux into a Distributed OS. He obtained a bachelors degree in computer science from Karnatak University in India in 1996.
Host: Xiaosong Ma, Computer Science, NCSU