| PrimeurWeekly 07 December 2009 |
|
|
 |
| |
| Special |
|
|
| How to make supercomputers a thousand times faster |
|
The fastest supercomputer in the world has a performance of about 1.7 Pflop/s. That's fast, very fast. It is Oak Ridge National Laboratory's Jaguar system. It is number one in the TOP500 of the world's fastest supercomputers. The number 500 in that list has a performance of about 20 Tflop/s. Still pretty impressive, but about 100 times slower that the number 1. On avarage, according to Jack Dongarra who spoke at the 25th anniversary supercomputer day at SARA in Amsterdam, it takes a number 1 system about 6-8 years to fall of the TOP500 list. So a number one machine gives you about 6-8 years a competitive advantage over organisations that do not have a supercomputer. The number 1 organisation can do science and engineering simulations that others cannot. According to Jack Dongarra the growth of computing power of the supercomputers will continue in the coming years and we will have the first computers wit Eflop/s performance somewhere around 2020. Problem is not to build these very large computers, but to program them: current software systems cannot handle these types of systems yet, and no-one knows if and how they can be programmed.
Read further... |
|
|
|
| Now we have Pflop/s systems, let's make them broadly usable |
|
Currently there are two supercomputers in the world with a performance of more than 1 Pflop/s (Petaflops), and even these two machines only are able to run so fast for only a limited set of applications. The Blue Waters project financed by the U.S. National Science Foundation is aimed at having sustainable Pflop/s performance by 2011. Deputy Director Bill Kramer explained at the 25th anniversary Supercomputer day at SARA in Amsterdam that reaching sustained Pflop/s not only requires designing and building the right hardware - what they do together with IBM, but also having the right operating system and middleware, and porting and tuning applications - what they do together with a number of user communities.
Read further... |
|
|
|
Advertisement
|
|
Advertisement
|
|
| |
| EuroFlash |
|
|
| EELA-2 project connects OurGrid to gLite over EDGeS Bridge |
|
The EELA-2 project has created a Bridge that connects the OurGrid middleware to gLite, the middleware that is also used by the large European EGEE Grid. This OurGrid-gLite Bridge is an implementation of the EDGeS 3G Bridge. Thanks to the new bridge, scientists in Latin-America can now seamlessly submit jobs to both Grids. When they do not have enough computing capacity on the gLite part of the EELA-2 e-infrastructure they can use the OurGrid resources.
Read further... |
|
|
|
| KnowARC releases Web Service based Grid middleware |
|
The KnowARC project has launched the Nox release of the Advanced Resource Connector (ARC) Grid middleware. This release brings ARC's new Web Service (WS) based components to a broader audience.
Read further... |
|
|
|
| Cray launches Exascale Research Initiative in Europe |
|
Cray has launched its Exascale Research Initiative at the Cray Executive Forum Europe currently taking place in Frankfurt, Germany. This research initiative will explore new ideas and technologies for overcoming the challenges of delivering a supercomputing system capable of sustained exaflop (Eflop/s) application performance.
Read further... |
|
|
|
| Peugeot Sport selects HyperWorks Suite to streamline development processes of LMP1 race car |
|
Peugeot Sport, the racing department of the Peugeot brand, has chosen Altair Engineering Inc.'s HyperWorks Suite to optimize composite development processes.
Read further... |
|
|
|
| Croatian broadband service provider optimizes its business with IBM |
|
Iskon, a Croatian Telco provider on the alternative broadband market, has selected technology from IBM to optimise its data back-up and archive system. The new IBM blade server and storage virtualization platform, which replaces the previously installed HP system, will help Iskon to remove system bottlenecks, streamline internal processes and help ensure uninterrupted service for customers. The system will also have the capacity to support Iskon's business growth, enabling it to introduce new services securely, quickly and in-line with increases in available bandwidth.
Read further... |
|
|
|
| JetBrains releases TeamCity 5.0 |
|
JetBrains have made available TeamCity 5.0, the company's award-winning distributed build management and continuous integration tool. TeamCity offers easy set-up and delivers out-of-the-box continuous unit testing, sophisticated code quality analysis, and early reporting on build problems. It has a gentle learning curve, so you can quickly improve your release management practices by gradually adopting its advanced features and capabilities.
Read further... |
|
|
|
| PRACE Award 2010: Call for Papers has Started |
|
The call for papers for the PRACE Award 2009 has begun. It will be the third time the PRACE Award will be presented at the ISC - International Supercomputing Conference - Opening Session.
Read further... |
|
|
|
| ISC honours five outstanding individuals as ISC Fellows |
|
Five prominent individuals from the HPC scene, Dr. Frank Baetke, Prof. Dr. Satoshi Matsuoka, Prof. Dr. Wolfgang E. Nagel, Prof. Dr. Horst Simon and Prof. Dr. Thomas Sterling have been honoured as ISC Fellows in the spirit of international co-operation, exchange of ideas and in recognition of their dedication and belief in the need for the yearly International Supercomputing Conference and Exhibition.
Read further... |
|
|
|
| New software to simulate future financial crises |
|
Can economics better predict how banks will react to future credit crunches and their impact on the wider economy? Breakthrough simulation software by European researchers could hold the answers to this question and more. How will economic policies adapt in 2020 when a quarter of the European Union population is over 65 and natural resources are dwindling? Can economists better predict future banking crises or economic turmoil? This week, the European Commission unveiled breakthrough research that could help answer questions like these by using economic simulation software.
Read further... |
|
|
|
Advertisement
|
|
| |
| USFlash |
|
|
| Futuristic Intel chip could reshape how computers are built, consumers interact with their PCs and personal devices |
|
Researchers from Intel Labs demonstrated an experimental, 48-core Intel processor, or "single-chip Cloud computer" that rethinks many of the approaches used in today's designs for laptops, PCs and servers. This futuristic chip boasts about 10 to 20 times the processing engines inside today's most popular Intel Core-branded processors. The long-term research goal is to add incredible scaling features to future computers that spur entirely new software applications and human-machine interfaces. The company plans to engage industry and academia next year by sharing 100 or more of these experimental chips for hands-on research in developing new software applications and programming models.
Read further... |
|
|
|
| Indiana University receives $1.5 million from NIH to explore Cloud computing for use in health research |
|
Researchers from the Pervasive Technology Institute Digital Science Center at Indiana University (IU) have begun work on an innovative project that will use Cloud computing techniques to support life science research. The project is supported by a $1.5 million grant award from the National Institutes of Health (NIH) and takes advantage of an earlier National Science Foundation grant to IU to construct an experimental supercomputing network called FutureGrid.
Read further... |
|
|
|
| Scale Computing announces ICS 2.0, adding data replication and snapshots |
|
Scale Computing has released its Intelligent Clustered Storage (ICS) architecture version 2.0. This second release addresses the three "C's" of IT storage purchasing requirements: cost, control and convenience. ICS 2.0 reduces costs, increases control and makes storage management more convenient.
Read further... |
|
|
|
| IBM expands ecosystem of partners to provide robust data centre solutions |
|
IBM has signed new and extended partnerships aimed at meeting the rising demand for energy efficient data centre capacity worldwide, while providing more flexible options for clients to meet business demand and IT growth. Speaking before an audience of over 1500 at Gartner's 28th Annual Data Center Conference, Steven Sams, vice president of Site and Facilities for IBM Global Technology Services, called to action the global leading power, cooling and cabling providers to partner with IBM and each other in new ways to drive cost optimization.
Read further... |
|
|
|
| Syracuse University, IBM and New York State launch one of the world's greenest data centres |
|
IBM, Syracuse University, with partners IBM and New York State, celebrated the construction of its new Green Data Center (GDC) - a showcase of world-class innovations in advanced energy-efficient information technology and building systems.
Read further... |
|
|
|
| University of Pittsburgh's McGowan Institute receives IBM Shared University Research Award |
|
The University of Pittsburgh's McGowan Institute for Regenerative Medicine, which connects scientists who work on tissue engineering, cell therapies, artificial organs and biodevices, has been awarded a Shared University Research (SUR) Award from IBM. The IBM SUR Award will provide IBM computer hardware, software and services to Pitt for its cutting-edge "in silico" biological research, which uses computer simulations to explore biological pathways and test therapeutic interventions.
Read further... |
|
|
|
| IBM takes top spot in server hardware revenue share in third quarter 2009 |
|
IBM again held the lead as the top server vendor worldwide in 3Q09 with 31.7 percent revenue share - compared with 30.2 percent for its nearest competitor - according to a new Gartner report.
Read further... |
|
|
|
| Researchers demonstrate a better way for computers to 'see' |
|
Taking inspiration from genetic screening techniques, researchers from Harvard and MIT have demonstrated a way to build better artificial visual systems with the help of low-cost, high-performance gaming hardware. The neural processing involved in visually recognizing even the simplest object in a natural environment is profound - and profoundly difficult to mimic. Neuroscientists have made broad advances in understanding the visual system, but much of the inner workings of biologically-based systems remain a mystery.
Read further... |
|
|
|
| Oracle Database on Fujitsu Server Cluster delivers world-record result on SAP Business Intelligence-Data Mart Standard Application Benchmark |
|
Oracle set a world-record result on the SAP Business Intelligence-Data Mart (BI-D) Standard Application Benchmark, surpassing the best IBM DB2 result on this benchmark with more than six times the performance.
Read further... |
|
|
|
| Sun announces Solaris Cluster 3.2 11/09 software |
|
Sun Microsystems Inc. has made available Solaris Cluster 3.2 11/09 software, the most innovative, cost-effective high availability platform based on the Solaris Operating System (OS). The latest release of Solaris Cluster enables increased use of the cost-saving, built-in virtualization features of the Solaris OS, allowing customers to run Oracle E-Business Suite, Siebel CRM 8, Single Instance Oracle Database and Oracle Real Application Clusters (RAC) inside Solaris Containers. It also provides greater consolidation options for end-to-end high availability and disaster recovery solutions with Oracle applications and databases.
Read further... |
|
|
|
| New Sun VirtualBox release enables first cross-platform "Teleport" of running virtual machine |
|
Sun Microsystems Inc. has launched a significant new version of Sun VirtualBox, its high performance, cross-platform virtualization software. VirtualBox 3.1 introduces the virtualization industry's first "Teleportation" capability, allowing running virtual machines to be moved, uninterrupted between disparate hosts - including those on different operating systems, different classes of computer (e.g. server to client) and even different CPUs (e.g. Intel to AMD).
Read further... |
|
|
|
| Verizon Business helps enterprises harness the power of the 'Cloud' |
|
Cloud computing is transforming how IT resources are delivered. For today's enterprises, it is not a question of if they will adopt "Cloud", or Internet-based, services, but how and when. Verizon Business is helping enterprises down this critical path with its new Cloud Computing Programme - a set of IT consulting and managed services that offer enterprises a holistic approach to transitioning to the Cloud, while addressing specific security, privacy, performance and governance requirements. The Verizon Cloud Computing Programme also helps enterprises manage, monitor and maintain their Clouds, whether applications are delivered from an enterprise's own data centre, or one operated by Verizon or a third party.
Read further... |
|
|
|
| Cray unveils "Cash-for-Clusters" buy-back programme |
|
Out with the old and in with the new. Cray Inc. has launched its own take on the United States government's recent "cash-for-clunkers" programme by launching a new buy-back programme for installed or used clusters or workstations. Cray's "cash-for-clusters" programme allows workstation and cluster users to trade in their current systems for credit towards the purchase of a new Cray supercomputer.
Read further... |
|
|
|
| A superbright supernova that's the first of its kind |
|
An extraordinarily bright, extraordinarily long-lasting supernova named SN 2007bi, snagged in a search by a robotic telescope, turns out to be the first example of the kind of stars that first populated the Universe. The superbright supernova occurred in a nearby dwarf galaxy, a kind of galaxy that's common but has been little studied until now, and the unusual supernova could be the first of many such events soon to be discovered.
Read further... |
|
|
|
| HP unveils Converged Infrastructure solutions for Microsoft environments |
|
HP has launched new integrations of the recently introduced HP Converged Infrastructure architecture for Microsoft applications, enabling customers to simplify the management of their computing environments while increasing efficiency.
Read further... |
|
|
