Tokyo Institute of Technology Gordon Bell Prize Winner Uses Bright Cluster Manager to Develop Applications for the TSUBAME 2.0 Supercomputer

Bright Computing today announced that Dr. Akinori Yamanaka, a Gordon Bell Prize winner and an assistant professor at the Tokyo Institute of Technology, relies on Bright Cluster Manager® to manage his personal cluster. Dr. Yamanaka writes and tests applications before running them on the institute’s TSUBAME 2.0 supercomputer— one of the world’s largest systems— designed to reach a sustained performance of 2.4 petaflops using a mixture of technologies from Intel, HP and NVIDIA.

The Gordon Bell Prize, considered to be the industry's highest honor, typically recognizes achievements by researchers using parallel computing to push the scientific envelope. Dr. Yamanaka and his colleagues were awarded the coveted prize in 2011 for their paper titled "Peta-scale Phase-Field Simulation for Dendritic Solidification on the TSUBAME 2.0 Supercomputer."

Dr. Yamanaka continues to research the mechanism and characteristics of binary metal solidification processes. He uses his personal cluster, built by Bright partner HPCTech Corp., as a development environment for a phase-field simulation program, for deployment on the TSUBAME 2.0 supercomputer. Phase-field simulation, involving the modeling of complex dendritic structures, is extremely compute-intensive. Dr. Yamanaka’s simulations utilize 16,000 CPU cores and 4,000 GPUs.

HPCTech built Dr. Yamanaka’s personal cluster to replicate the TSUBAME 2.0 system, to ease his efforts in developing scalable programs for deployment on TSUBAME 2.0. As importantly, Dr. Yamanaka was looking for a cluster management solution that would require minimum effort, and no learning curve, so that he could focus on his research. To meet this requirement, HPCTech built Dr. Yamanaka’s system with Bright Cluster Manager.

“We scientists are time-constrained,” said Dr. Yamanaka. “Our priority is our research, not managing our clusters. Bright is intuitive to use, and with it I can effectively manage my cluster without wasting time writing scripts or synchronizing management tool revisions. Provisioning is fast and easy too. I prefer this approach over open source toolkits.”

“Bright is particularly beneficial for program developers– especially CUDA developers,” said Hiroshi Sato, an Engineering Manager at HPCTech. “Bright provides an excellent, comprehensive development environment for CUDA and HPC, and completely integrates GPU management into its management solution. Bright gives scientists like Dr. Yamanaka complete visibility and control of their cluster, presented in an intuitive GUI. Bright takes the complexity out of cluster monitoring and management.”

For more information:

Bright Cluster Manager: http://bit.ly/x8NIOG
CUDA Development Environment: http://bit.ly/BrightGPU
Provisioning: http://bit.ly/Provisioning

About HPCTech Corp.
HPCTech, an authorized reseller of Bright Cluster Manager, is a high-performance computing solution provider for hardware and software, as well as a systems integrator for scientific computing. Founded in 2008, HPCTech is the sole agency of Innodisk/ACTICA memory in Japan. http://www.hpctech.co.jp/

About Bright Computing
Bright Computing specializes in management software for clusters, grids and clouds, including compute, storage, Hadoop and database clusters. Bright’s fundamental approach and intuitive interface makes cluster management easy, while providing powerful and complete management capabilities for increasing productivity. Bright Cluster Manager is the solution of choice for many research institutes, universities, and companies across the world, and manages several Top500 installations. Bright Computing has its headquarters in San Jose, California. http://www.brightcomputing.com

Mark Blessing
Bright Computing
2880 Zanker Road
San Jose, CA 95134
www.brightcomputing.com
mark.blessing@brightcomputing.com

This release was published on openPR.