HPC NEWS
Contents
HPC News
The latest news in
high-performance computing
HPC Products
A brief update of the latest HPC software and hardware
Unbounded clusters
Paul Schreier on the growing complexities of cluster management
22 20
18
ISC10 announces keynote speakers
The 25th International Supercomputing Conference, ISC10, will feature four keynote presentations on a wide spectrum of topics, from scaling up and scaling out of HPC servers, to the use of HPC in China. ISC takes place from 30 May to 3 June in Hamburg, and comprises a packed conference and more than 140 exhibitors.
Visualising HPC
Stephen Mounsey on
the visualisation tasks requiring HPC
26
Kirk Skaugen, vice president of the Intel Architecture Group and general manager of the Data Center Group, Intel, will discuss trends that will help accelerate deployment (of affordable HPC systems) in the server market. The industry manufacturing HPC systems was and still is focused on high-end systems and is now aiming at the exascale boundary. On the other hand, the affordability of systems in the terascale range now makes ‘scale out’ (spreading the HPC computing base) even more possible than before.
Professor Dr Helmut Merkel
HPC Director 30
Lee Ward, principal
member of technical staff, Sandia National Laboratories
of EurAsia Global Concept in Peking, China, will provide an insight into ‘How the IT/ Internet Revolution Changes the Chinese Society’. There are 380 million Internet users in China, making it the world’s biggest online community. Professor Dr Thomas
Sterling, the Arnaud and Edwards Professor of Computer Science at Louisiana State University, will look at the beginning of the 2 AP (after petaflops) era, which promises to be an exciting and a dramatic leap in technology. New multicore chip architectures, the building blocks of many of the prominent supercomputers worldwide, will be identified and described to track the growth and trends in performance opportunity. The presentation will conclude with the ‘Canonical HPC System’, a design point that represents the more widely used components, sources, and scales over the preceding year.
Cray to work with Microsoft Research on cloud computing
Cray’s custom engineering group will work with Microsoft Research to explore and prototype a system that could provide a glimpse into the future of cloud computing infrastructure.
The objective is to design a supercomputing architecture that dramatically lowers the total cost of ownership for cloud computing data centres. Cray’s custom engineering group will design a system infrastructure that combines super efficient power delivery, high-density packaging and innovative cooling technologies. This solution is intended to reduce facility, power and hardware costs significantly. ‘This is an ideal project for our custom engineering group,’ said Chuck Morreale, Cray’s vice president of custom engineering. Christian Belady, director of hardware architecture in Microsoft’s Extreme Computing Group added: ‘Ultimately, our mission is to reduce total cost of ownership dramatically, while significantly boosting performance.’
Danish university eases wireless communication transceiver simulation
Cover montage Dean Farrow
18
Aalborg University in Denmark is using GPU technology to enhance its work in the theory, modelling, design, implementation and test of RF systems and circuits. Aalborg’s Department of Electronics Systems and Technology Platforms Section has made significant effort in the use of GPUs to solve numerical problems in the communication technology area, in order to improve communication systems and devices.
SCIENTIFIC COMPUTING WORLD APRIL/MAY 2010
As a stepping stone towards
broader use of GPU technology, Professor Torben Larsen’s team at AAU has employed a number of systems housing various GPU configurations to explore the possibilities. The researchers’ primary interest is the development of improved algorithms and techniques that lead to useful discovery, which is why they have chosen to develop in a very high-level language (VHLL). The algorithms
will be developed on these systems using Jacket (from AccelerEyes) and Matlab. The development of new algorithms that incorporate Jacket from the beginning means that there is not a ‘transformation’ process to get GPU results. Upon completion of the development, users can either run plain Matlab on CPU-based systems or leverage the above GPU resources to gain performance advantages instantly.
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 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44