July 14, 2011
Advanced HPC cluster with Bright Cluster Manager used to help unravel the mysteries of the universe and the world of molecular dynamics
SAN JOSE, Calif., July 15 -- Bright Computing, a leader in cluster management software, today announced that Drexel University is using Bright Cluster Manager to manage its newly-installed HPC cluster from Advanced HPC. This system, funded by an NSF grant, is being used to perform research at two extremes of science: simulations of star-forming regions in space, including galactic nuclei, and molecular dynamics. As far apart as these studies appear, they are both based on similar techniques employing particle methods. In addition to research, the cluster is also used for teaching courses in high-performance computing and GPU programming.
Astrophysicist Dr. Steve McMillan is the Principal Investigator using the cluster. There are three co-investigators from multiple departments at Drexel: Dr. Cameron Abrams, Chemical and Biological Engineering; Dr. Jeremy Johnson, Computer Science; and Dr. Nagarajan Kandasamy, Electrical and Computer Engineering.
The team's cluster is currently the most powerful compute resource at Drexel University, capable of delivering 176,514 GFLOPS of GPU performance. The system comprises 68,352 NVIDIA GPU cores and 48 TB RAID disk storage.
"I am impressed with the capabilities of the Advanced HPC team," said Dr. McMillan. "They took the time to truly understand our requirements, and designed a system that was even better than our original specifications. We have an incredibly efficient cluster that exceeds our expectations, and has given us more bang for the buck."
Bright Cluster Manager is used for remote provisioning, monitoring and managing the system. Dr. McMillan and his team were amongst the first to use Bright Cluster Manager revision 5.2, with its full support of NVIDIA's CUDA 4.0 and associated metrics.
"Bright Cluster Manager was a good choice for us," stated Dr. McMillan. "I'm a scientist, not a system manager. The Bright GUI is intuitive and extremely easy to use. I can do everything I need to do on the system remotely -- I almost never visit our cluster. I get alerts when attention is required, and am able to hand off much of the overall system management with confidence to an enthusiastic undergraduate."
"From a support perspective, Advanced HPC and Bright are a great team," added Dr. McMillan. "Initially we experienced a problem with some of our jobs, and they responded immediately. As it turned out, the issue was a driver problem. The Bright people worked directly with NVIDIA to solve it -- sparing us the pain and getting us productive again in a timely fashion."
About Advanced HPC, Inc.
Advanced HPC is a provider of HPC platforms and Linux clusters. Their product scope spans Linux cluster solutions, high-performance servers and workstations, disk storage products such as Storage Area Networks (SAN), Network Attached Storage (NAS), and Direct Attached Storage (DAS), tape backup solutions from external tape drive to enterprise-class libraries, disk-to-disk-to-tape solutions and data deduplication solutions, backup software and virtualization solutions. Advanced HPC's team has over 50 years of experience in the technology industry and their staff includes industry veterans in sales, customer service and production to ensure that customers get the best solution to fit their needs. http://www.advancedhpc.com
About Bright Computing
Bright Computing is a specialist in cluster management software for high-performance computing (HPC). Its flagship product -- Bright Cluster Manager -- with its intuitive graphical user interface and powerful cluster management shell, makes clusters of any size easy to install, use and manage. Bright's minimal footprint enables HPC systems to be utilized to their maximum potential, from departmental clusters to large-scale systems. Bright Cluster Manager is the management solution of choice for many research institutes, universities, and companies across the world, including several TOP500 installations. Bright Cluster Manager is certified as Intel Cluster Ready. Bright Computing has its headquarters in San Jose, California. http://www.brightcomputing.com
Source: Bright Computing, Inc.
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?