September 26, 2013
XSEDE (the eXtreme Science and Engineering Discovery Environment) is a virtual organization, funded by the National Science Foundation, that supports advanced research with a collection of local and national digital resources and services. Today XSEDE is launching a new program – the campus bridging initiative – that aims to unify these resources, making it easier for researchers to go back and forth between local clusters and national supercomputing resources.
As Dr. Craig Stewart of Indiana University explains "the idea is to let an individual student or a researcher or a staff member treat local cyber-infrastructure, local computers and national infrastructure all as if they were just peripherals attached to the back of your laptop."
As part of the campus bridging initiative, XSEDE is rolling out new software that lets researchers and IT managers build a "basic XSEDE-compatible computing cluster" from scratch. The service is based on cluster automation software called "Rocks Roll."
In a video overview of the project, Dr. Craig Stewart of Indiana University and Dr. Marcus Alfred of Howard University discuss XSEDE's goal to use the campus bridging program as a means of making research easier. The XSEDE Compatible Cluster Kit, a Rocks Rolls distribution, paves the way for researchers and admins to set up an XSEDE compatible resource at their own institutions.
Dr. Stewart explains the impetus for campus bridging. "If you were a researcher used to using your own campus resources and you went to use one of the national facilities, it felt like you were falling off a cliff," he says. "The transfer of knowledge from local systems to nationally-funded systems was very low and the user had to do a lot of relearning. Campus bridging is focused on making it easier for researchers to transfer their skills from their local clusters to the big clusters that are in XSEDE."
The Rocks Rolls distribution is just the first of many goals associated with the campus bridging project. Next, XSEDE will release libraries of RPMs, to facilitate the transfer of tools for the basic XSEDE-compatible cluster build. RPMs will also be used as a cluster maintenance tool. Not only does the new initiative close the gap between local clusters and national supercomputers, it facilitates consistency. Skills attained on XSEDE clusters will transfer to campus clusters and vice versa – documentation and instructional material will likewise be streamlined.
"This is a significant step in our campus bridging efforts," notes John Towns, principal investigator and leader of XSEDE. "The new tools are designed to ensure that the national cyberinfrastructure ecosystem we are developing and growing supports a balanced approach to the use of national resources such as XSEDE and campus-based resources in a way that accelerates American research efforts and maximizes US global competitiveness."
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?