October 08, 2012
SAN JOSE, Calif. Oct. 8 — Bright Computing – the leading vendor-independent provider of cluster management software – announced Brazil's prestigious National Institute for Space Research (Instituto Nacional de Pesquisas Espaciais – INPE) chose Bright Cluster Manager to manage computing clusters in its Laboratory for Computing and Applied Mathematics (LAC). The center, focused on applied research for aerospace engineering, uses the HPC cluster to simulate spacecraft control and orbit guidance. INPE selected Bright Cluster Manager because its unified solution for provisioning, job scheduling, monitoring and management reduced both complexity and workload for the center's systems administrators. Further, Bright makes it easy to use its integrated MPI and mathematical libraries without specialized training.
The National Institute for Space Research's LAC cluster is Brazil's first computer based on the Intel E5-2620 "Sandy Bridge" processor, which provided another point of validation for INPE's choice of Bright: Intel's HPC development team used Bright throughout the Sandy Bridge development process, ensuring strong compatibility between the hardware and the cluster management solution. Bright partner Scherm Brasil, in conjunction with 12 research scientists from multiple academic departments, built and installed the new cluster in a single day.
"Bright Cluster Manager is a powerful tool for building and testing HPC clusters," said Guilherme Friol, CTO at Scherm Brasil. "We installed Bright on bare metal in under an hour, including our choice of workload manager that Bright provides as a pre-configured option. With Bright, we were able to quickly configure and test the system far faster than what is possible with other cluster management software. We also felt good about providing our customer with Bright for ongoing cluster management. They are impressed with Bright, and therefore pleased with us."
The LAC professors use a multitude of data libraries and models including several libraries provided by Bright:Hierarchical Data Format (HDF5), Portable Extensible Toolkit for Scientific Computation (PETSc), Intel Integrated Performance Primitives (IIPP), and the GNU MultiPrecision library (GMP). The scientists needed to quickly access and manage compilers, the message passing interface (MPI) libraries, and the mathematical libraries. Bright Cluster Manager makes it fast and easy for the researchers to access and use multiple libraries and compilers.
"We needed a solution that minimized the time and expense of cluster management. With Bright, we are able to spend nearly all of our time simulating and investigating complex aerospace engineering models," said Dr. Elbert E.N. Macau, professor of Computing and Applied Mathematics. "We chose Bright because its highly intuitive, integrated approach for managing our HPC cluster and associated libraries saves us a lot of time, and eliminates the need to hire staff to manage our cluster."
Source: Bright Computing
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?