Primeur magazine

Edition: live - Issue: 2012-06-20

Hardware

Gilad Shainer from Mellanox Technologies presented his research to look into exascale matters. HPC has begun scaling beyond the Petaflop range towards the Exaflop mark. The computing world is changing and the existing and ever growing demand for performance creates a need for megaclusters. Read further...
Julian Kunkel from the German Climate Center talked about the simulation of parallel programmes on the level of applications and systems. He tried to give an answer to some pending questions. What performance do we expect for running an MPI-IO application on an arbitrary real cluster? Where are the bottlenecks during the execution? Is the observed performance of an MPI call optional? Read further...

Applications

NVIDIA's GPUs will be used by scientists at Germany's Forschungszentrum Jülich, which hosts the Jülich Supercomputing Centre, one of Europe's largest and most powerful supercomputing resources, to accelerate advanced neurological research targeted at unlocking secrets of the human brain. Read further...
NVIDIA's GPUs are playing an integral role in research that's shedding new light on ways to prevent future epidemics of the deadly H1N1 influenza virus. A major difficulty in combating H1N1 has been the frequent and rapid onset of new virus mutations that rendered anti-influenza drugs, such as Tamiflu (oseltamivir) and Relenza (zanamivir), ineffective. Read further...

TOP500

Pavan Balaji from Virginia Tech expanded on the Green500 List and the escapades to Exascale. One of the goals of the Green500 List is to drive energy efficiency as a first-order design constraint, on par with FLOPS. He emphasized the need to encourage a fair use of the list rankings. Read further...

Company news

Mellanox has launched Connect-IB, a scalable server and storage adapter solution for High-Performance Computing (HPC), Web 2.0, Cloud, Big Data, financial services, virtualized data centres and storage environments. Connect-IB adapters deliver the highest throughput of 100Gb/s utilizing PCI Express 3.0 x16, unmatched scaling with innovative transport services, sub-microsecond latency and 130 million messages per second - 4X higher message rate over competing solutions. Read further...
Mellanox continued its commanding lead as the global interconnect solution provider for the TOP500 list of supercomputers with InfiniBand now leading as the top interconnect of choice for the TOP500 with 210 clusters. From June 2011 to June 2012, the total number of InfiniBand-connected CPU cores on the TOP500 list grew 44 percent, 86 percent of accelerator-based systems are connected with InfiniBand, and the amount of InfiniBand-based system performance grew 70 percent. This growth highlights the increasing demand for InfiniBand as a way to maximize computing and accelerator resources, productivity and scalable performance in the world's fastest computer systems. Read further...
The scientists at the National Chemical Laboratory (NCL) in India have selected Bright Cluster Manager to help streamline the process of installing, integrating and managing new clusters for polymer-based research. These scientists - who run complex molecular dynamics simulations on HPC clusters - previously relied on scripting Rocks with other open source tool kits to manage their clusters. The NCR team chose Bright instead, to eliminate the need for Linux expertise or custom scripting. NCL's scientists can focus on their research instead of wasting time synchronizing management toolkits and squinting at endless command lines to keep their clusters running. Read further...
Terascala Inc. has introduced LustreStack, the industry's first software suite and open framework for speeding the development and deployment of Lustre-based storage appliances on industry-standard hardware based on Intel Xeon processors. Read further...
Rutherford Appleton Laboratory (RAL), one of the UK's principal government labs and part of the Science and Technology Facilities Council (STFC), has deployed a large-scale Ethernet network to support a new high performance computing (HPC) facility that has 4.5 petabytes of storage on-line. Read further...
The SAGA system, India's most powerful supercomputer and the holder of the 86 position on the Top500 list, is leveraging NVIDIA GPUs to dramatically improve the design and analysis of the delivery vehicles critical to the nation's space programme. Read further...
NVIDIA Tesla K10 GPUs offer performance breakthroughs on popular high performance computing (HPC) applications - ranging from seismic processing to life sciences to video processing - according to new benchmarks NVIDIA. Read further...