news and products
HIGH-PERFORMANCE COMPUTING
Diary Dates
21-23 January 2013 8th International Conference on High-Performance and Embedded Architectures and Compilers (HiPEAC) Berlin, Germany
21-23 January Sixth Workshop on Programmability Issues for Heterogeneous Multicores (MULTIPROG-2013) Berlin, Germany
18-21 March GPU Technology Conference 2013
San Jose McEnery Convention Center, California, USA
• To ensure your event is listed, please send details to
editor.scw@
europascience.com
he 40th edition of the twice-yearly Top500 list of the world’s leading supercomputers has been
released, and early indications that Oak Ridge National Laboratory’s Cray XK7 system, Titan, would take the top spot proved accurate. Boasting 560,640 processors, including 261,632 Nvidia K20x accelerator cores, Titan achieved 17.59 petaflops on the Linpack benchmark, beating the record of 16.32 petaflops previously set by Sequoia, an IBM BlueGene/Q system installed at Lawrence Livermore National Laboratory. Now the number two system, Sequoia – at 1,572,864 cores – is the first system with one million or more cores. The third spot on the list has been claimed by Fujitsu’s K computer installed at the RIKEN Advanced
Institute for Computational Science (AICS) in Kobe, Japan; the fourth position goes to a BlueGene/Q system named Mira at Argonne National Laboratory; and Juqueen, a BlueGene/Q system, at the Forschungszentrum Juelich in Germany completes the top five. Following an upgrade, Juqueen is now the most powerful system in Europe. Another new system in the Top 10 is Stampede, a Dell PowerEdge C8220 system installed at the Texas Advanced Computing Center at the University of Texas in Austin. It uses the new Intel Xeon Phi processors (previously known as MIC) to achieve its 2.6 petaflops. In total, 23 systems on this latest list have achieved petaflops performance. It should be noted that, despite delivering petascale
For regular news updates, please visit
www.scientific-computing.com/news
Titan tops Top500 T
NEWS IN BRIEF
Record-setting simulations at DOE laboratories exploit the power of IBM Blue Gene/Q supercomputers Intel has endorsed Green Revolution Cooling’s CarnotJet submersion cooling system following one year of testing Appro International, a
developer of advanced scalable supercomputing solutions, has been acquired by Cray
New products XC30
Cray has launched its next-generation high-end supercomputing system, the XC30. Previously code- named ‘Cascade’, the Cray XC30 supercomputer is what the company describes as its most advanced high- performance computing (HPC) system ever built. Combining the new Aries interconnect, the Intel Xeon processor
E5-2600 product family, Cray’s fully-integrated software environment and innovative power and cooling technologies, the XC30 is designed to scale HPC workloads of more than 100 petaflops.
Early shipments of the Cray XC30
are starting now and systems are expected to be widely available in the first quarter of 2013. Future versions of the XC family of supercomputers will be available with the new Intel Xeon Phi coprocessors and Nvidia Tesla GPUs based on the next-generation Nvidia Kepler GPU computing architecture.
www.cray.com
12 SCIENTIFIC COMPUTING WORLD
performance on applications, the Cray Blue Waters system at NCSA at the University of Illinois at Urbana-Champaign, chose not to submit a Linpack benchmark performance figure.
For regular product updates, please visit
www.hpcprojects.com/products
Allinea DDT for IBM Blue Gene/Q
Allinea Software has announced the release of its high-performance scalable parallel debugger, Allinea DDT, for IBM BlueGene/Q systems. Argonne National Laboratory (ANL), host to Mira, an IBM BG/Q capable of carrying out 10 quadrillion calculations per second on more than 750,000 cores, has worked closely with Allinea for the past two years to adapt Allinea DDT’s scalable architecture to the older IBM BlueGene/P and to prepare for the IBM BlueGene/Q architecture. This work has seen Allinea DDT set
debugging scale and performance records for this system family. ‘This tool has already proven its value in the migration of our early science applications onto Mira,’ said Kalyan Kumaran, who manages the Argonne Leadership Computing Facility’s applications performance engineering team. ‘These projects cover the range of scientific fields, numerical methods, programming models and computational approaches expected to run on Mira, so accurate debugging is critical.’
www.allinea.com
www.scientific-computing.com
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32