January 02, 2013
Jan. 2 – ASUS ESC4000/FDR G2 HPC servers have been chosen to meet the needs of the SANAM HPC cluster used by the Frankfurt Institute for Advanced Studies (FIAS) in Germany and the King Abdulaziz City for Science and Technology (KACST) in Saudi Arabia to support national-level scientific research. As of November 2012, SANAM was number two on the global Green500 list, which ranks supercomputers by energy efficiency, making it the most energy-conserving GPU-based supercomputer in the world. It stood at number 52 on the worldwide TOP500 supercomputer list, with 421.2TFLOPS total processing power. SANAM employs 210 ASUS ESC4000/FDR G2 computing nodes with 420 AMD FirePro S10000 dual-GPU modules and an array of 420 Intel Xeon E5-2650 8-core processors for a total of 38,400 cores.
Power efficient hybrid computing performance
The ESC4000/FDR G2 was picked for SANAM supercomputer usage in fields such as petroleum, aerospace, bioinformatics, weather forecasting, and the simulation of chemical reactions and particle movement. It delivers the latest HPC technology while offering superior parallel processing and extra-fast low latency InfiniBand FDR data transfers at up to 56Gbps. Its architecture makes it an excellent choice for heavy-duty calculations and modeling-intensive applications, with the overall ASUS ESC range of products already well-established in finance, virtualization, and high-end scientific uses. The Green500 list ranked SANAM number two in November 2012 for sustaining highly efficient performance at 2.3GFLOPS per watt, or 2.3 billion calculations per second per watt. It ranked number 52 on the TOP500 list in November, reaching 421.2TFLOPS at half the cost of a similarly-powered CPU-only design. SANAM benefits from ESC4000/FDR G2 hybrid computing architecture, which makes the most of combined CPU/GPU parallel processing.
ESC4000/FDR G2 offers greater expandability
The ESC4000/FDR G2 presents a 2U hybrid CPU/GPU supercomputer/HPC server. Its flexible design enables greater choice, supporting Intel Xeon Phi coprocessors and GPUs from AMD and NVIDIA. It can also accommodate dual Intel Xeon E5-2600 processors on the board at once, and assures high expandability via nine PCI Express 3.0 x16 slots. For cooling, intelligent controls manage CPU and GPU-area fans individually, and to provide easy maintenance, all eight 3.5” hard drive bays utilize a downtime-minimizing hot swap design. ESC4000/FDR G2 energy efficiency gains from the integration of twin 1620W 80 PLUS Platinum power supplies, which are up to 94% efficient to promote energy savings.
Rick Rozalsky, Vice President of Global Business Development, Adtech Global:
"When building the SANAM supercomputer, we focused on utilizing technologies that provide continuously improving performance, lower costs, and reduced power consumption in the future. We chose ASUS to satisfy our need for a flexible, global systems manufacturer that would offer us the engineering and development depth needed for this advanced heterogeneous platform. ASUS not only met our specifications and criteria, but we also saw potential for them to assist in BIOS-level development, mechanical design flexibility, and other future product enhancements. ASUS acted as a true partner in every aspect of execution and product supply."
Tom Lin, Server Business Unit General Manager, ASUS:
“ASUS is honored to take part in the SANAM project with Adtech Global to utilize our extreme performance and optimized supercomputers in scientific research. ASUS brings its own research and development expertise to the table, and is committed to providing the best designs for scientific development, research, and applications.”
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?