HPC: ISC’10 SHOW PREVIEW
Hamburg highlights
A brief preview of ISC’10, which runs from 31 May to 4 June in Hamburg and features an exhibition, with many new products being launched
Pavan Balaji, an assistant computer
scientist at Argonne National Laboratory,
will be presenting a tutorial on high- speed networks. The tutorial will provide an introduction to InfiniBand (IB) and 10-Gigabit Ethernet (10GE) interconnects, their offered features, their current market standing, and their suitability for prime- time HPC. It will start with a brief overview of IB and 10GE and their architectural features.
In addition, the emerging OpenFabrics stack will be discussed, which encapsulates both IB and 10GE in a unified manner, and hardware technologies such as Virtual Protocol Interconnect, remote direct memory access (RDMA) over Ethernet, and RDMA over Converged Enhanced Ethernet, which aim at converged hardware solutions. IB and 10GE hardware and software solutions and the market trends will be highlighted. The tutorial will also include sample performance numbers illustrating the effectiveness that these technologies can achieve in different environments such as MPI, sockets, parallel file systems, multi-tier data centres, and virtual machines.
www.anl.gov
Bull will be
exhibiting its newly extended and updated family of servers designed for extreme computing: Bullx. HPC users have extremely diverse
requirements, and are increasingly searching for hybrid solutions to cover the widest possible spread of applications. Bull has recognised this, and has been gradually building an exhaustive range of solutions that includes thin nodes (Bullx blade
40
system), fat nodes (Bullx supernodes) and accelerators (Bullx accelerator blades with integrated GPUs). They can all be combined with each other in a customised way, and managed as a single system using the bullx cluster suite. The Bullx blades and supernodes have been designed and developed entirely by Bull’s HPC teams, Europe’s largest team of experts. In fewer than five years, Bull has won more than 150 customers in 15 countries across three continents. Many prestigious companies and research centres have chosen Bull Extreme Computing solutions, most notably in France, Germany, UK, Spain and Brazil.
www.bull.com
Convey Computer will demonstrate its hybrid-core computing system, the Convey HC-1. The HC-1 combines advanced architecture and compiler technology with off-the-shelf x86 and FPGA processors, helping customers accelerate application performance while reducing energy, floor space, and management costs. Hybrid-core computing has ‘personalities’, which are reloadable instruction sets implemented in hardware that run on the Convey coprocessor. This technology dramatically increases the performance and efficiency of applications by executing critical regions of code on the reconfigurable Convey coprocessor. Personalities can feature a series of instructions that accelerate general HPC applications or be specific to individual algorithms.
www.conveycomputer.com
Cray will showcase its comprehensive line of HPC systems, including the latest additions to the Cray XT line of massively parallel processing supercomputers and the Cray CX family of supercomputing systems. The Cray XT systems combine sustained
SCIENTIFIC COMPUTING WORLD JUNE/JULY 2010
application performance with exceptional manageability and reliability to create highly scalable Linux supercomputers. The systems span thousands to millions of processing cores using powerful AMD Opteron processors and according to Cray are the first general purpose, scalable petascale supercomputers.
The Cray CX family of Intel-based
cluster systems includes affordably-priced supercomputers that are easy to configure, deploy, administer and use. ‘Right-sized’ in performance, functionality and cost, Cray CX systems are designed for a wide range of users, from the single user who wants the power of a cluster next to the desk, to a department of users who want a shared cluster resource.
www.cray.com
The German Climate Computing Centre
(DKRZ), the worldwide national service centre supporting climate system research, will exhibit at the event. DKRZ provides high-performance computing platforms, sophisticated and high-capacity data management, and services for climate science.
DKRZ runs an IBM Power6 system with a peak performance of more than 150 teraflops/s. It is used for complex climate model calculations, which contribute to the IPCC assessment reports. DKRZ will present some of the results with the spherical display, Omniglobe – a virtual planet with a 360° view.
www.dkrz.de
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 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60