July 03, 2006
Arizona State University's
(ASU) Fulton High-Performance Computing (HPC) Initiative Center
provides researchers faster access to collaborative research
information vital to solving complex scientific and engineering
problems due to its adoption of the Panasas ActiveScale Storage
Cluster. By deploying the Panasas ActiveScale Storage Cluster, ASU
has increased its Linux cluster performance by 10 times while eliminating
the need for costly NFS servers.
The HPC Initiative serves as a university hub for high-performance
computing. ASU researchers use the facility's 200-node Linux cluster to
study complex problems ranging from the design of golf balls that fly
farther and straighter, to developing massively large numerical
simulations that can lead to more accurate weather predictions. In
building out the HPC Initiative Center, ASU recognized that reliance of
legacy storage solutions would require multiple NFS servers to scale,
realized this was not a viable solution, and selected Panasas to meet
their increased performance requirements and their scalability needs.
"We evaluated several storage solutions with parallel file systems,
knowing that a reliance on legacy NFS servers would not enable us to
scale quickly or match the demands of our Linux cluster. Over time,
adding more NFS servers would stifle the cluster's usability," said Dan Stanzione, director of the High Performance Computing Initiative.
"In evaluation tests on 32- and 64-node test data, Panasas with
DirectFLOW outpaced competitive storage solutions right
out-of-the-box, without requiring any fine-tuning. Quite simply, we
plugged in Panasas storage and it worked. It's reliable and remarkably
The Panasas Storage Cluster with DirectFLOW provides a
fully-parallel data path for high speed and direct communications
between the facility's Linux cluster and Panasas storage, eliminating
performance bottlenecks that can idle the cluster as it waits for data.
Instead, Panasas storage optimizes the speed in which meaningful
results can be processed and delivered.
In addition, Panasas' integrated hardware/software solution
streamlines management tasks so that IT managers can be more productive
and spend less time on tedious administrative tasks. The Panasas
Storage Cluster also supports clustered NFS so that non-Linux nodes can
access data running on the Linux cluster. This means that ASU faculty
and students running applications from their Windows-based workstations
can use the Panasas Storage Cluster to process their jobs directly on
the Linux cluster. In this way, the facility can fully maximize IT
investments in its infrastructure and reduce the overall cost of
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?