November 25, 2005
The Pacific Northwest National Laboratory team showed its storage muscle, winning the StorCloud competition at the Supercomputing 2005 conference last week in Seattle.
To demonstrate its capabilities, PNNL streamed video data from high-capacity storage on the conference show floor to Hewlett-Packard-loaned hardware in the laboratory's booth. The amount of data being moved ranged from 30 to 60 gigabits per second. At the low end, this is the equivalent of transferring and processing a full DVD of video every second.
The direct access to data on parallel file systems across high-speed wide area links was made possible by recent network bandwidth increases at PNNL, including new fiber optic connections to Richland and a connection to the Department of Energy's UltraScience Net exchange in Seattle. UltraScience Net provides on-demand dedicated bandwidth channels at multi, single, and sub lambda resolutions. PNNL used six dedicated 10 gigabit lambdas to transfer data from an HP StorageWorks Scalable File Share (HP SFS) parallel file system at SC05 to an Itanium cluster located more than 250 miles away. HP SFS is based on Lustre Technology from Cluster File Systems Inc.
"Researchers often have data stored at one site, access to computational resources at another site, and are physically located at a third site, " said Kevin Regimbal, manager of high performance computing at PNNL. "The traditional solution to this problem is for a user to copy data to a desired location, which could take several days, perform calculations, and then copy the results back to the first location for post-processing and analysis. Thanks to the UltraScience Net, the National Lambda Rail and other research networks, we're starting to approach speeds that could allow researchers to access their files directly, regardless of locality."
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?