Primeur live 2012-06-19

Hardware

Stanford engineers perfecting carbon nanotubes for highly energy-efficient computing ...

CSC, CSCS, SARA, and T-Platforms to deliver scalable hybrid prototype for PRACE ...

Global HPC scientific research, development and innovation at the ISC'12 exhibition floor ...

TOP500

Graph500: if you cannot measure it, you cannot improve it ...

How to run the Linpack benchmark for even more powerful HPC systems to come ...

Company news

Allinea Software visualizes the future ...

Cray awarded $16 million contract from the Department of Defense High Performance Computing Modernization Programme ...

CSC - IT Center for Science Ltd. selects SGI ICE Cube Air to increase HPC capacity ...

Organisations convert Big Data into a business driver with solutions from NetApp ...

Businesses to take advantage of big data analytics with enterprise-ready Apache Hadoop solutions from NetApp and Hortonworks ...

Adaptive Computing integrates key updates into its Moab HPC Suite, helping customers get the most from HPC ...

Mellanox enables Peer-to-Peer communication between GPUs with new NVIDIA GPUDirect technology ...

Mellanox FDR 56Gb/s InfiniBand to accelerate world-class supercomputer at TACC ...

CSIRO Australia selects Mellanox end-to-end FDR InfiniBand interconnect solutions for new compute infrastructure ...

Mellanox 40GbE interconnect solutions optimize HPC application performance for Ethernet-based server and storage clusters ...

Cray to add Intel Xeon Phi coprocessors to its next-generation "Cascade" supercomputer ...

Whamcloud and Swiss National Supercomputing Centre sign Lustre support agreement ...

First commercial IBM hot-water cooled supercomputer to consume 40% less energy ...

Primeur magazine

Edition: live - Issue: 2012-06-19

Hardware

Energy efficiency is the most significant challenge standing in the way of continued miniaturization of electronic systems, and miniaturization is the principal driver of the semiconductor industry. "As we approach the ultimate limits of Moore's Law, however, silicon will have to be replaced in order to miniaturize further", stated Jeffrey Bokor, deputy director for science at the Molecular Foundry at the Lawrence Berkeley National Laboratory and Professor at UC-Berkeley. Read further...
CSC - IT Center for Science Ltd., CSCS - Swiss National Supercomputing Centre, SARA - Dutch national centre for HPC & e-science support, and T-Platforms have signed a partnership to deliver a next-generation supercomputer prototype system. The system is part of the future HPC technology assessment activities of the pan-European PRACE Research Infrastructure (PRACE RI). Read further...
The ISC'12 exhibition is buzzing and running smoothly now. The ISC'12 team organized a research tour across the showfloor to provide an overview of what is cooking in the academic world in the HPC domain. We can assure you: it is a lot and it is very global. We were taken to India, China, Japan, Europe and the United States. Each academic booth responsible had a remarkable story to tell. Read further...

TOP500

At ISC'12, Richard Murphy from Sandia National Laboratories, held a presentation on the philosophy, challenges and state of the Graph500 list. Graph50 is an attempt to measure for an emerging critical problem domain: the fundamental mathematical model of relationships. This is difficult for the computer to solve. Graph problems tend to exhibit a lot of data access and very little compute work. They have the potential for huge impact to national security and the commercial world. The goal is to use Graph500 to ease the modern computing communication bottleneck. Read further...
Jack Dongarra from the University of Tennessee and Oak Ridge National Laboratory presented a solution to keep the Linpack benchmark run time manageable for future TOP500 lists. As more and more very powerful supercomputers are entering the list, run times are exceeding human limits. Read further...

Company news

Allinea Software will provide delegates of the International Supercomputing Conference (ISC12) with a preview of the latest version of Allinea DDT 3.2, a debugging tool for High Performance Computing. New additions to the feature set provide ground breaking abilities to enhance the speed at which bugs can be fixed. Read further...
The Department of Defense High Performance Computing Modernization Programme (DOD HPCMP) has awarded a contract to Cray to significantly upgrade the programme's three existing Cray XE6 supercomputers. As part of the contract, Cray will also deliver a new Cray XE6m supercomputer to the Naval Research Laboratory in Monterey, California, which will be used by the organisation's Marine Meteorology Division. Read further...
Finland's CSC - IT Center for Science Ltd. (CSC) has selected SGI ICE Cube modular data centre technology for a new green data centre in Kajaani, Northern Finland. Valued at 2.6 million euro, the three year deal will see a brownfield site transformed to meet CSC's expanding demand for HPC capacity. Read further...
Organisations across a wide range of industries are dealing with the realities that big data presents. Machine- and user-generated data growth is exploding in size and expected to increase at a rate of 40% year over year. Because organisations must be able to quickly and efficiently turn this data into valuable information to drive smarter business decisions, they require a storage solution that delivers continuous data availability for optimal performance levels and simplified management. Read further...
Customers have gained two new choices to deploy enterprise-ready Apache Hadoop solutions and solve their big data analytics requirements. First, NetApp and Hortonworks entered a strategic partnership to develop and pretest joint Apache Hadoop solutions using Hortonworks Data Platform. These solutions will help customers simplify access to their data and gain deeper business insight to make the right decisions faster. Second, NetApp unveiled the NetApp Open Solution for Hadoop Rack, which combines network, compute, and storage resources for customers who need a ready-to-deploy enterprise Hadoop solution. The new solution bolsters NetApp's existing Hadoop portfolio and gives customers greater choice and flexibility to deploy Hadoop in their enterprise. Read further...
Adaptive Computing, manager of large supercomputing systems and expert in intelligent HPC workload management software, has introduced enhancements to its Moab HPC Suite that improve workload management and accelerate high performance computing. The Moab HPC Suite 7.1 now offers faster processing of parallel and big data jobs, as well as streamlined scheduling and visualization workload capabilities that improve the overall productivity of HPC systems. Read further...
Mellanox Technologies will support the new generation of NVIDIA GPUDirect technology. NVIDIA GPUDirect technology dramatically accelerates communications between GPUs by providing a direct peer-to-peer communication data path between Mellanox's scalable HPC adapters and NVIDIA GPUs, without transferring data to the CPU or server memory subsystem. Read further...
Mellanox Technologies Ltd.'s FDR 56Gb/s InfiniBand solutions will provide the high-performance interconnect enabling extremely scalable parallel applications for the new Texas Advanced Computing Center (TACC) Stampede supercomputer. Read further...
The Commonwealth Scientific and Industrial Research Organisation (CSIRO), Australia's national science agency, has selected Mellanox's end-to-end FDR InfiniBand solution for their new accelerated cluster, including SwitchX SX6025 switches, ConnectX-3 adapters, Unified Fabric Manager (UFM) and cables. CSIRO chose Mellanox's robust, high-speed interconnect solution due to its scalability, low power consumption, and high-efficiency data handling that utilizes the latest GPUDirect and CORE-Direct acceleration technologies. Read further...
Mellanox Techologies' 40 Gigabit Ethernet interconnect solutions with RoCE (RDMA over Converged Ethernet) support have been optimized to deliver the highest performance for compute and storage intensive applications. Mellanox is the first and only company to deliver a complete end-to-end, RoCE enabled, 40GbE solution including NICs, switches, cables and software accelerators. Mellanox 40GbE based clusters, accelerated with RoCE technology, deliver a more than 80 percent application performance increase compared to 10GbE based clusters. For storage access, Mellanox 40GbE delivers 4X faster storage throughput, enabling new levels of storage density and dramatic savings in CAPEX and OPEX. Read further...
Cray Inc.'s next-generation supercomputer code-named "Cascade" will be available with Intel Corporation's new Intel Xeon Phi coprocessors. The Intel Xeon Phi product family is based on Intel's Many Integrated Core (Intel MIC) architecture, which is designed for highly parallel workloads. Read further...
Whamcloud, a venture-backed company formed by a worldwide network of high-performance computing (HPC) storage industry veterans, has signed a Lustre support agreement with the Swiss National Supercomputing Centre (CSCS). Read further...
The Leibniz Supercomputing Centre (LRZ), in collaboration with IBM, has introduced the world's first commercially available hot-water cooled supercomputer, a powerful, high-performance system designed to help researchers and industrial institutions across Europe investigate and solve some of the world's most daunting scientific challenges. Read further...

Primeur live 2012-06-19

Hardware

Stanford engineers perfecting carbon nanotubes for highly energy-efficient computing ...

CSC, CSCS, SARA, and T-Platforms to deliver scalable hybrid prototype for PRACE ...

Global HPC scientific research, development and innovation at the ISC'12 exhibition floor ...

TOP500

Graph500: if you cannot measure it, you cannot improve it ...

How to run the Linpack benchmark for even more powerful HPC systems to come ...

Company news

Allinea Software visualizes the future ...

Cray awarded $16 million contract from the Department of Defense High Performance Computing Modernization Programme ...

CSC - IT Center for Science Ltd. selects SGI ICE Cube Air to increase HPC capacity ...

Organisations convert Big Data into a business driver with solutions from NetApp ...

Businesses to take advantage of big data analytics with enterprise-ready Apache Hadoop solutions from NetApp and Hortonworks ...

Adaptive Computing integrates key updates into its Moab HPC Suite, helping customers get the most from HPC ...

Mellanox enables Peer-to-Peer communication between GPUs with new NVIDIA GPUDirect technology ...

Mellanox FDR 56Gb/s InfiniBand to accelerate world-class supercomputer at TACC ...

CSIRO Australia selects Mellanox end-to-end FDR InfiniBand interconnect solutions for new compute infrastructure ...

Mellanox 40GbE interconnect solutions optimize HPC application performance for Ethernet-based server and storage clusters ...

Cray to add Intel Xeon Phi coprocessors to its next-generation "Cascade" supercomputer ...

Whamcloud and Swiss National Supercomputing Centre sign Lustre support agreement ...

First commercial IBM hot-water cooled supercomputer to consume 40% less energy ...