high-performance computing Te company will also be showcasing its latest
technologies including a live preview of its HPC Burst Buffer and DDN’s Exascale Architecture – Infinite Memory EngineTM (IME).
www.ddn.com
CloudFlow – Computational Cloud Services and Worlflows for Agile Engineering – will enable the remote use of computational services distributed on the cloud, seamlessly integrating these within established engineering design workflows and standards.
www.eu-cloudflow.eu
water for cooling, thus boosting energy performance by around 40 per cent compared to traditional data centers, despite being just as easy to maintain as standard, air-cooled servers. Te Bullx DLC B700 range is based on industry- standard components, which are all cooled through direct contact with a cold plate.
www.bull.com./extreme-computing
CoolIT’s RACK DCLC product range features scalable liquid cooling solutions that meet even the most aggressive density, efficiency and performance
requirements. Integrating the rack-based direct contact liquid cooling technology into your data center will reduce overall energy consumption, increase CPU/GPU density, and provide higher performance per core. Visit us at Booth #411 to discover how to get 45kW+ per rack with warm water cooling.
www.coolitsystems.com
DataDirect Networks (DDN) is a provider of massively scalable storage, and powers over two- thirds of the systems in the Top 100. DDN will discuss storage challenges around
ingesting, processing, storing and distributing big data and also invites delegates to learn more about the company’s innovations in driving the highest levels of performance for storage environments, scalability to meet today’s (and tomorrow’s) demanding requirements and process more within tighter budgets. DDN will hold its User Group meeting on the
benefits of cache-centric storage technologies in tackling large-scale storage and I/O challenges on Tuesday 24 June at 7.30am in the Multi-Purpose Area 2 Floor 1.
www.scientific-computing.com l
DKRZ is a national service centre for climate researchers who depend on HPC resources. At ISC’14, DKRZ will give a preview of its upcoming high performance computer HLRE-3
On 6 May 2014, DKRZ and Bull signed a
contract for the delivery of a petaflop-scale supercomputer in 2015/2016, an agreement was also reached to provide HPC support for climate research. Te contract covers the delivery of all the
key computing and storage components. Furthermore, DKRZ would also like to inform attendees about its data services including data documentation, quality assurance, long-term archiving and international network data access. DKRZ computers are regularly used for the
extensive simulations whose results contribute to the IPCC reports. DKRZ’s climate globe – a virtual planet with a 360-degree view – is used to present visualisations of the latest of these climate simulations.
www.dkrz.de
Edinburgh Parallel Computing Centre (EPCC) researches and develops novel computing solutions; writes soſtware;
manages computing and data systems, and provides HPC and data-related training; it also hosts Archer, the UK’s national HPC service. At this year’s ISC, the EPCC will focus
on a few specific project areas, including Fortissimo, which looks to give European manufacturing (particularly SMEs) greater efficiency and competitive advantage through the use of simulation services running on cloud infrastructure making use of HPC systems. Te Adept project, addressing the challenge of
the energy-efficient use of parallel technologies by building on the expertise of HPC soſtware developers to exploit parallelism for performance, and on the knowledge of embedded systems engineers to manage energy usage.
www.epcc.ed.ac.uk
Fabriscale is a start-up that specialises in fabric management soſtware with an emphasis on smart algorithms that simplify network configuration, management and routing. At ISC’14, it will announce its first product, the
@scwmagazine
During the show GiDEL will be presenting new architecture based on
FPGAs to enable dramatically reduced power consumption while increasing computing performance. Te FPGA’s inherent low-power dissipation
achieves unmatched power utilisation. Moreover, architecture based on massive parallel connectivity reduces power and latency while sustaining cluster scalability without degrading unit performance. With the development of the OpenCL for FPGA this platform-independent language enables more programmers to utilise the power of FPGA based architecture.
www.gidel.com
Fabriscale Fabric Manager. In the first release of the Fabriscale Fabric Manager for InfiniBand, the centre of attention is fast fault-tolerance. When the network fails, the consequences can be dramatic, but with Fabriscale users are protected against the negative consequences of any switch or link failure independent of network topology. Any incidents are handled automatically, transparently, and quickly by the Fabriscale routing engine. Come visit at ISC’14 in booth number 751
for a live demonstration of InfiniBand fabric management with fast fault-tolerance.
www.fabriscale.com
➤
In today’s top scientific research world, the computing acceleration with GPUs and MIC cards are a must. While combining regular CPUs
with several computing cards in a same enclosure designed to deliver high performance computing power, Gigabyte still delivers high density, power efficiency, and excellent cooling performance. With its innovating solution designed
for scientific analysis and simulation, computational finance, weather & climate forecasting, and oil and gas simulation. Te GS-R22PHL offers eight double slot GPU/MIC cards when the market supports six cards at best in similar 2U enclosures.
b2b.gigabyte.com/products/ product-page.aspx?pid=4498#ov
JUNE/JULY 2014 37
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