The ISC exhibition enables HPC vendors to connect with 2,500 buyers and decision-makers within three days. ISC has a proven track record in attracting academic and business leaders from around 50 countries. With the move to Leipzig, ISC’13 offers the opportunity to be a part of the largest and most modern HPC exhibition in Europe.
Adaptive Computing (Booth #500) will show how technologies from Adaptive and NICE Software are making technical visualisation workload optimisation (TVWO) a reality for IT. They will explain how through the corporate data centre and Adaptive’s collaborative efforts, it’s possible to build a full-blown, physical-server-based Private Technical Compute Cloud. What Saas did for 2D applications – keeping applications, compute resources, and data together in the data centre where they can be accessed by PCs, thin clients, and mobile devices – can now be accomplished with complex 3D technical simulations.
Benefits of TVWO include: increased workforce collaboration and productivity; 3D applications closer to your data; compute capabilities ‘right-sized’ on the fly, utilisation maximisation; less network congestion, fewer bottlenecks; centralised and shared servers are more efficiently utilised; and hardware and software are more efficient, less costly, and don’t affect users as much as when completed at individuals’ desks.
Advania (#601) is a Nordic IT company that operates throughout the Scandinavian region. Advania has a staff of 1,100 people and 20 offices throughout the Nordic region. The company has given more than 70 years of information technology service to both the private and public sectors.
Advania operations are certified according to international quality and security standards ISO 9001 and ISO 27001. The company offers data centre services ranging from basic colocation, cloud services, data storage and completely managed services in green and emissions free Icelandic Tier 3 data centres.
Today, the company has more than 50 international clients using its data centre. One of its biggest clients is NHCP, the Nordic Supercomputing Project (http://nhpc.hi.is/). NHCP chose to locate to the Advania data centre due to the low energy price and an infrastructure that can handle high density environments.
Allinea Software (#655) is a vendor of tools for parallel software development, offering a complete toolset for parallel debugging and profiling. This unified platform provides an easy-to-use environment for: each stage of development, from concept to production; every platform, with native clients for OS/X, Windows and Linux; and systems such as Cuda, Xeon Phi, MPI, OpenMP, UPC, and OpenACC.
Powerful diagnostic capability, intuitive graphical interface and lightning performance – even on the largest systems – combine to make Allinea DDT a preferred tool for users in research, academia and industry in the US, Europe and Asia.
Allinea MAP is an MPI profiler that works without slowing down a program. It shows users which lines of code are slow and gets everything else out of the way. Allinea MAP works out of the box, whether at one process or 10,000, with just five per cent overhead.
Altair (#266) offers best-in-class application software, HPC workload management tools and a licensing/business model for growing computing needs. The company’s PBS Works suite of HPC solutions includes PBS Professional, which optimises the world’s largest supercomputers. It is the workload management software of choice by many supercomputing sites within the Top500, providing security, reliability and energy-efficiency.
Web-based Compute Manager allows users to run, monitor, and manage workloads on distributed resources. Its results and remote visualisation capabilities enable working with big data efficiently and securely. Display Manager enables seamless remote visualisation of big data, preventing large data movement across networks for optimised performance. PBS Analytics, a job accounting/reporting solution, provides administrators with advanced analytics to support data-driven planning.
Altair will also introduce private cloud appliance, HyerWorks Unlimited. This is a state-of-the-art private cloud solution with fully configured hardware and software, offering unlimited use of all Altair software within the appliance.
Asetek (#271) is a leading provider of energy-efficient liquid cooling systems for data centres, servers, workstations, gaming and high-performance PCs. Its products are used for reducing power and greenhouse emissions, lowering acoustic noise, and achieving maximum performance by leading OEMs and channel partners around the globe.
Highlights at ISC'13 will include its RackCDU system which provides data centres with cooling cost reductions exceeding 50 per cent, density increases of 2.5 times, and the ability to recycle up to 80 per cent of their IT power.
At this year’s ISC, Boston (#263) will be proudly unveiling its ARM-as-a-Service, the world's first ARM Cloud server dedicated to facilitating software migration to the ARM architecture. The ARM-as-a-Service (AaaS) has been designed specifically to ease the transition of moving or migrating software to the ARM architecture. The solution is based on the Boston Viridis Microserver, the world's first self-contained, ultra-low power cluster solution that is based on processor technology from ARM.
AaaS is the world’s first commercially available Cloud offering, based on Calxeda EnergyCore ARM-based processor technology. AaaS is a platform which provides developers with all the tools and services required to port and migrate software to the ARM platform. Built specifically to assist in migrating and porting applications from x86 to ARM, Boston has teamed up with software developers Ellexus for their product Breeze, a choice of product also made by ARM to profile and troubleshoot applications on its own HPC cluster.
At ISC’13, Bull (#245) will be showing its broad range of Extreme Computing solutions, with a special focus on the bullx B700 Direct Liquid Cooling blade series, which delivers drastic savings in energy consumption. Cooling takes place inside the blade itself, through direct contact between the heat-producing components and a cold plate with coolant circulating within it. This enables warm water to be used for cooling, boosting energy performance by around 40 per cent compared to traditional data centres, despite being just as easy to maintain as standard, air-cooled servers.
Bull will also be exhibiting a full range of accelerated servers. And for HPC users who are worried about the scalability of their codes on many-core systems, experts from the recently inaugurated Bull Center for Excellence in Parallel Programming will be present and ready to answer questions and suggest solutions.
Chelsio (#502) is focused on solving high performance networking and storage challenges for virtualised enterprise data centres, cloud service installations, and cluster computing environments.
With its fifth generation protocol acceleration technology, it is delivering hardware and software solutions including Terminator 5-10/40Gb Ethernet Unified Wire network adapter cards and unified storage software. With its proven Terminator ASIC technology designed in more than 100 OEM platforms and the successful deployment of more than 100,000 ports, Chelsio has enabled unified wire solutions for LAN, SAN and cluster traffic.
The company has taken the unified wire to the next level with full offload for TCP, UDP, iSCSI, iWARP RDMA and Fibre Channel over Ethernet. Network convergence over 10Gb Ethernet is happening now and Chelsio is a driving force behind the consolidation.
Ciara Technologies (#361) is a global leader in the design, development and integration of HPC solutions. Its Nexxus C deskside cluster now includes the addition of liquid cooled Intel Xeon Phi coprocessors into the GPU node. Each node can support up to four Intel Xeon Phi coprocessors or an impressive 4.5 teraflops.
In November 2011, Ciara launched the Nexxus C deskside cluster. This revolutionary design can support up to 20 processors, 16 GP-GPUs or Intel Xeon Phi coprocessors and nearly 2TB of memory. It is designed to work in a lab or office workspace with acoustics in mind due to the utilisation of an industrial-grade liquid cooling solution.
The addition of the Xeon Phi cards represents a significant paradigm shift in HPC. Developers comfortable with traditional x86 compilers can now enter into the HPC space to develop massively parallel applications. Ciara is at the forefront by integrating liquid-cooled cards into its deskside cluster.
ClusterVision (#520) specialises in the design, build and management of high-performance compute clusters. At ISC’13, ClusterVision, together with its technology partners, Green Revolution Cooling, Supermicro and Intel, will demonstrate a working prototype of a new fully submerged server cooling system. This innovative solution provides effective and energy efficient heat dissipation.
ClusterVision will also be sharing the ISC’13 Exhibitor Forum stage with The University of Paderborn to describe their new Top500 class cluster. The 200 Tflops system, located at the Paderborn Center for Parallel Computing (PC²), contains 614 compute nodes with 10,000 cores, and incorporates the latest GPU and many-core acceleration technologies available. It will run a range of commercial and open source HPC applications and will be used for fundamental research and development projects in computational design and performance. Exhibitor Forum 02, Tuesday, 18 June: 11am-11.30am.
Coolcentric and Eaton-Williams (#661) have teamed up to deliver what they say are the most energy and space efficient cooling solutions. The Coolcentric family of heat exchangers comprise passive, liquid cooled heat exchangers; standard RDHx and low-density RDHx-LD to replace standard rear doors on IT rack enclosures. Close-coupled to the IT enclosure, heat exchangers bring cooling as close to the heat source as possible, providing the ultimate containment solution. Coolcentric heat exchangers are reliable, scalable, energy efficient and space-saving turn-key cooling solutions that help drive down capital and operation costs. Coolcentric heat exchangers can provide sensible cooling for rack loads from 5kW to 40kW.
The Eaton-Williams Coolant Distribution Units monitor and manage the flow of cooled, treated water in a closed loop environment to the Heat Exchanger units. Factory certified technicians are available to provide installation, commissioning and preventive maintenance throughout the world.
CoolIT Systems (#301) has been dedicated to the invention and design of state-of-the-art liquid cooling technology since 2001. It is an experienced innovator with 43 patents and more than one million liquid cooling units deployed in desktop computers, servers and data centres around the world.
This year at ISC’13, in booth 301 as well as a number of partner booths, CoolIT features various rack-based direct contact liquid cooling (Rack DCLC) solutions, including the AHx (Air Heat Exchanger) and the CHX (Coolant Heat Exchanger). CoolIT is also hosting a BoF session, ‘Building Liquid Cooling Technology Standards’, on 18 June.
CPU 24/7 (#600), based in Potsdam, Berlin, is highly specified in providing high-performance computing (HPC) systems and computing power ‘on demand’ for industry and universities, for applications in development and research. With many years of experience and expertise, CPU 24/7 recognises the need for specific computing power resources, e.g. in the field of numerical simulations or physical processes such as flows, burning and chemical reactions.
CPU 24/7 provides these HPC resources in a timely manner, at a reasonable price and using innovative and the latest technology. At ISC, CPU 24/7 will present its two ready-to-work packages: Resource Area as online bookable, flexible-use and immediately available HPC Cluster, and the Tailored Configurations as rentable, continuously usable and individually planned and customised HPC clusters. Both are remote and on-demand. New customers are eligible for a special ISC offer.
DataDirect Networks (DDN) (#305) is the world's largest privately held storage company and supplies the storage behind more than 60 per cent of the world’s fastest supercomputers. DDN will hold its European User Group on the morning of 17 June. This will include talks from customers who have faced significant Big Data challenges, but are now leveraging the fastest storage systems in the world to solve toughest HPC challenges in high-energy physics, chemistry, CFD, weather, materials, energy life science and more.
DDN will be also be showcasing its latest technologies including the SFA12K, hScaler, and the new 7700 hybrid flash storage appliance.
Eurotech (#306) is a HPC manufacturer and solutions provider. Visitors can discovery how the company can help organisations to crack the problem of many ambitious goals constrained by lack of time and resources. At ISC’13, Eurotech will display some products from the Aurora line of supercomputers like the direct hot water cooled Aurora Tigon, capable of impressive performance (350 Tflops/rack) and energy efficiency (measured > 3200 Mflops/Watt on 64 nodes). The new and compelling Aurora G-Station, the silent water cooled mini HPC ideal for wherever there is no data centre infrastructure, and the cluster of accelerators developed in the D.E.E.P. project, will also be presented.
Extoll (#660) is a European developer and vendor of HPC network technology. Extoll network technology comprises low-latency, direct networks with each node having up to six links and an integrated switch. At ISC’13, the company will announce the start of production of the Extoll-Asic (Tourmalet) which will be available in Q4 2013. According to Extoll, Tourmalet will provide extremely low latency (<0.6µs), ultra-high message rates (>100 Mill. msg/sec), and a link bandwidth of >10GB/s.
Tourmalet will be the interconnect for the Booster part of the European DEEP project. For early adopters, the Extoll interconnect technology is available today on FPGA (Galiber for PCI-express; Ventoux for HyperTransport). At the booth, Extoll will showcase live demos of direct GPU to GPU communication without any CPU involvement. It is also proud to announce mass production capacities for its high-density Active Optical Cable (HD-AOC), to be used for cluster interconnections of up to 25m.
GiDEL (#560), a bronze sponsor at ISC’13, was founded in 1993 as a high-end system development and integration company. With a project-level approach, it created several powerful and advanced tools for high-performance system development.
Leading manufacturers, OEMs and researchers use GiDEL reconfigurable FPGA-based PROC boards to accelerate real-time processing in financial analytics, data mining, and molecular dynamics. They are also used for accelerating the processing of large number crunching batch applications, such as bioinformatics, and seismic/exploration analysis. Highlights at ISC'13 include life science applications, DSP (digital signal processing) and HPRC (high performance reconfigurable computing), as well as high-speed low latency networking and network analysis, complex algorithm and IPs validation, ASIC and SoC prototyping. A presentation on Multi Dimension Direct Connectivity: Architecture & Applications will take place on Tuesday, 18 June at 3.45pm – 4.15pm in the exhibition hall, #290.
Gompute (#513) delivers comprehensive solutions for high-performance computing in-house, in-the-cloud or both. With more than 10 years' experience working with industrial clients, providing solutions and services to the IT and engineering communities, as well as advisory functions to C-level executives, it delivers a collaborative and highly productive work environment for geographically distributed engineering teams.
Combined with Gompute On-Demand, the company ensures that users have the option to either scale out or completely host their HPC environment, ranging from one to 1,000s of cores in a Gompute-owned supercomputing centre.
Megware (#644) is a German developer and vendor of high-performance computing solutions. The company’s focus is on customer‐oriented, tailor‐made and innovative system solutions and services. At ISC'13, Megware will showcase its in‐house developments, such as the hot water direct cooling solution, ColdCon. ColdCon is adapted to Intel and AMD processors and is suitable for high packaging densities due to the Megware SlashEight blade system. In addition to processors and chipsets, further hotspots, such as storage modules have been incorporated into the cooling system. Large scale saving of energy can be achieved by means of free re-cooling and by reusing the waste heat produced.
For the ISC Student Cluster Challenge, Megware supports the team ‘TurboTUC’ (Chemnitz University of Technology) (#494) by providing hardware and technical advice.
At ISC'13, Nvidia (#220) will be featuring advances in applications and scientific discovery made with GPU-accelerated computing. The Nvidia booth will showcase the latest GPU technologies in accelerated computing, including the Tesla Kepler-based GPU Accelerators, as well as taking a look at how GPUs are being used to advance scientific discoveries. Bill Dally, Nvidia’s chief scientist will kick-off this year’s conference with his opening keynote: Future Challenges of Large-Scale Computing. In addition, several sessions and tutorials will feature talks on software, tools and programming models using Nvidia Tesla GPUs.
Obsidian (#383) is demonstrating the new Longbow C400, a 4X QDR InfiniBand range extender capable of spanning continental distances with complete protocol transparency and local line rate performance using standard quad 10Gbits/s xWDM channel WAN links. Building on the legacy of Obsidian's first generation 4X SDR Longbow C100, the C400 adds advanced feature upgradability such as inter-subnet routing, line-rate AES cryptography, Forward Error Correction and precision WAN monitoring and characterisation functions. Obsidian is also featuring the A-CWDM81, which is a passive CWDM mux/demux that combines with the C400 to provide up to 80Gbits of InfiniBand traffic over a single fibre pair. Both devices comply with Obsidian's 1/4 rack unit form-factor for high density deployments at very low power levels.
Wide Area InfiniBand is very suitable for long distance, high-performance data migration, remote synchronous storage mirroring, hitless data centre expansion across multiple buildings, cluster aggregation and remote visualisation applications.
Panasas (#101) designs high-performance parallel network attached storage (NAS) solutions for technical computing applications and Big Data workloads. Panasas ActiveStor scale-out NAS appliances use the Panasas PanFS parallel file system to deliver superior performance, data protection, scalability and manageability. Panasas products are optimised for high-demand storage applications in bioscience, energy, government, finance, manufacturing, media, and other core research and development sectors.
At ISC’13, Panasas will be making a couple of announcements, one of which is the launch of a new software product that will benefit many applications including life sciences, energy exploration, and, in particular, media/entertainment. Other highlights will include demonstrations on how blade-based Panasas ActiveStor scale-out NAS appliances integrate the PanFS operating system to deliver superior performance and scalability, and how intelligent use of SSD and SATA technologies assures unmatched performance for mixed file size workloads.
Pentair’s (#007) global business unit, Pentair Equipment Protection, is a provider of product and service solutions for the protection and cooling of electrical and electronic systems. It supports the market demands of the fields of automation, control systems and communication in a wide spectrum. Its brands offer a comprehensive palette of off-the-shelf, modified and technical solutions for the trade, communications, energy, electronics, industrial, infrastructure, medical, and security and defence systems markets. Pentair Equipment Protection includes the Hoffman, McLean and Schroff brands.
Its brand Schroff offers a broad selection of standard, configured and individually adapted solutions for telecommunications and infrastructure. Its datacom solutions combine modular components and scalable, standard products. They all work together to create a solution attuned to individual requirements.
PRACE (#725) will start off its presence at ISC’13 with the PRACE Scientific Conference on Sunday, 16 June, featuring high-level scientific presentations. The PRACE booth at the ISC13 exhibition will feature the six leading-edge systems that it offers access to for scientists and researchers from academia and industry. The six systems are located in France (GENCI), Germany (GCS), Italy (CINECA) and Spain (BSC).
Themed presentations will provide visitors with detailed information on PRACE’s training offer, e.g. the PRACE Advanced Training Courses and the PRACE Seasonal Schools. PRACE will highlight its PRACE 2.0 Strategy for 2015 onward, which will take it into its next phase as a persistent pan-European research infrastructure. PRACE will also organise a BoF on PRACE services for Industry. Visitors are invited to come play the ‘Dare To Think The Impossible’ game: shoot some stars and win a T-shirt!
Q-Leap Networks (#562) will present the newest version of Qlustar, its full-fledged HPC cluster operating system (OS). It is the first and only professional Debian/Ubuntu based solution in the field providing many unique features. Qlustar is designed to scale from small departmental clusters to multi-thousand node supercomputers. Its core is a lightweight, modular Compute/Storage-Node OS. As a whole, it is a complete Linux distribution based on Debian/Ubuntu. QluMan, Qlustar's management and monitoring framework, handles easy and consistent cluster setup, keeping track of hard and software failures, cluster operation with ease, speed and control, and more.
With Qlustar, node provisioning belongs to the past: just assign an OS image to a node and (re)boot, and it will be ready for use in a few minutes. Hot fixes in an OS Image can easily be applied on-the-fly to a running node. Fully integrated support for Lustre storage clusters, GPU computing and high-availability configurations is available amongst others.
Rogue Wave Software (#550) is an independent provider of cross-platform software development tools and embedded components for the next generation of HPC applications. Offering a broad portfolio, Rogue Wave enables developers to increase productivity and harness the power of multicore computing while reducing the complexity of developing multi-processor and data-intensive applications.
TotalView is a scalable and intuitive debugger for parallel applications written in C, C++, and Fortran. Designed to improve developer productivity, TotalView simplifies and shortens the process of developing, debugging, and optimising complex applications. TotalView provides a powerful combination of capabilities for pinpointing and fixing hard-to-find bugs, such as race conditions, memory leaks, and memory overruns. Providing developers the ability to step forwards and backwards through program execution, TotalView’s reverse debugging capabilities reduce troubleshooting code. TotalView provides debugging support for Nvidia Cuda, OpenACC, and Intel Xeon Phi coprocessors.
RSC Group (#440), the Russian full cycle HPC solution provider, will demonstrate the RSC Tornado at ISC’13. RSC Tornado is the patented cutting-edge energy-efficient scalable architecture with direct liquid cooling on widely available commodity server boards (i.e. Intel S2600JF) with top Intel Xeon E5-2690 processors, Intel SSD and high-speed InfiniBand FDR interconnect. It provides the industry-record 180+ Tflops/rack performance with Intel Xeon Phi coprocessors, outstanding PUE 1.06 and 100+ kW/rack power density. Two Intel Xeon Phi projects, the first outside the USA, being deployed in Russia and highly ranked by the Top500 and Green500 lists, will be presented at the company’s booth.
Samplify (#365) is a software company accelerating memory, storage, and I/O bottlenecks in HPC, big data, big science, and cloud computing. The company’s APAX acceleration technology is a universal encoder/decoder for all numerical data, and imaging types. Samplify’s APAX HDF provides a three to eight times reduction in HDF file size for climate data, multi-physics, CFD, visualisation, and analytics.
Implemented as a plug-in to HDF, Samplify’s APAX requires no modification of solver codes to achieve the benefits of accelerating throughput to storage subsystems and expanding storage capacity. At ISC’13, visitors can attend a paper presentation by the University of Hamburg and DKRZ, benchmarking Samplify’s APAX technology on climate data. The paper presentation is scheduled for Wednesday, 19 June at 9.40am in Hall 5.
At ISC’13, Samsung (#200) is presenting its 20 nanometer class DRAM and latest solid state drives as well as DDR4 Dram for the new generation of high performance server memory. With its partners, Flytech and AdTech, Samsung is showcasing its Green Memory Solutions for data centre servers, virtualisation, high-performance computing, and big data.
With Christmann Informationstechnik + medien, Samsung is illustrating its Exynos application processor for HPC applications, which powers Christmann’s RECS | Box Compute Unit 3.0. The Exynos 5 Dual, based on the ARM Cortex-A15 CPU, has been designed with both performance and power efficiency in mind to offer an impressive user experience in mobile devices, and maximise its benefits in HPC environments.
Scalable Informatics (#551) is a provider of high performance, tightly-coupled storage and compute solutions for a wide range of data, I/O and process-intensive applications. Its innovative design — tightly coupling storage and computing in the same unit — eliminates bottlenecks to deliver the highest performance at aggressive price points. Scalable will be demoing JackRabbit at ISC, one of the fastest high performance single storage and computing devices on the market.
The tight coupling between processing and I/O lets customers run applications directly on JackRabbit at significantly higher application acceleration, delivering greater streaming data bandwidth, higher IOPs and I/O throughput, and more effective server utilisation for improved productivity and lower operational costs. The 4U Software Defined Storage, JR4-T60b, sustains more than seven Gbytes, scales to 240TB, and is usable with a multitude of networking protocols, block/file targets, and cluster file systems. In addition, JackRabbit has options for multiple ports of 1/10/40 GbE and IB.
ScaleMP (#262) will be showing the latest version of vSMP Foundation which is available as a free download. In addition, ScaleMP will consult visitors to create a large SMP system that the company says is 20 per cent less expensive with 20 per cent more compute power and 20 per cent more memory than its competitor can offer. vSMP Foundation allows for the creation of large scale symmetric multiprocessors (SMPs) from industry standard servers based on the latest Intel or AMD CPUs.
vSMP Foundation scales to more than 32,000 CPUs and 256TB of memory. As data and compute needs have grown significantly in recent years, vSMP Foundation has evolved to allow more data to be kept in memory which speeds up big analytics applications. In addition, with the ability to create SMPs as the workloads change, a software-defined data centre can be implemented which responds to users requirements.
For ISC’13, SGI will not be exhibiting at the main ISC conference venue, but will instead have meeting space and a demonstration room at nearby hotel, NH Hotel Leipzig Messe. Dr Eng Lim Goh, chief technology officer, senior executives, and technical experts from SGI will be present to discuss the company’s solutions, and define how they can accelerate a variety of workflows common amongst HPC customers.
SGI will be showcasing solutions that will allow researchers rapid access to their input, while at the same time eliminating data silos and driving up cost efficiency for IT organisations. Specifically this will include Scientific Big Data – UV 2000 Blade, Energy Efficiency with M-Cell – ICE-X Blade, Storage and Cloud Storage – Modular InfiinteStorage
Located within 300 metres walking distance of the CCH Leipzig, the venue for SGI is: NH Hotel Leipzig Messe, Fuggerstrasse 2, 04158 Leipzig.
Spectra Logic (#541) defines, designs and delivers innovative tape and disk-based exascale archive storage products for high-performance computing (HPC) organisations. To help HPC organisations push the boundaries of their research and exploration objectives, the company’s product features will help ensure data is always available with cost-effective storage that meets performance, growth and environmental requirements.
By working closely with customers, Spectra Logic has designed and delivered products that address archive-related pain points commonly experienced in high-performance compute environments. Visitors will receive a hands-on demonstration of the exabyte capacity T-Finity tape library.
Super Micro Computer (#320), known as Supermicro, is a global leader in high-performance, high-efficiency server technology and innovation, and is a premier provider of end-to-end green computing solutions for HPC, data centres, cloud computing, enterprise IT, Hadoop/big data and embedded systems worldwide. Supermicro's advanced server Building Block Solutions offer a vast array of components for building energy-efficient, application-optimised, computing solutions.
Products include SuperServer and SuperStorage solutions, FatTwin, MicroCloud, SuperBlade, GPU and Hyper-Speed systems, SuperWorkstations, motherboards, chassis, Platinum-Level high efficiency (95 per cent +) digital switching power supplies, Battery Backup Power (BBP) modules, 10/1GbE network switches, Server Management Software Suite and SuperRack cabinets/accessories delivering performance and value.
Founded in 1993 and headquartered in San Jose, California, Supermicro is committed to protecting the environment through its ‘We Keep IT Green’ initiative. The Company has operations centres in Silicon Valley, the Netherlands and its Asia Science & Technology Park and Logistics Center in Taiwan.
ttgLabs (#504) is an optimisation solutions provider that addresses the problem of software optimisation for GPUs. It has solid expertise in software autotuning for supercomputers and HPC systems of hybrid architecture (CPU+GPU) within MPI, OpenMP, CUDA, OpenCL, and OpenACC paradigms. At ISC’13 it will showcase a new version of its flagship product, TTG Apptimizer v1.0.3, which has been recently installed on ‘Lomonosov’ system, the No.26 in the Top500 list (Nov’2012).
This toolkit enables developers to optimise resource-demanding applications for GPU-based systems and to easily achieve an efficient usage of all processing units, dynamic load balancing, selection of the optimal computational cores, and detection of the required level of parallelism. New features advance optimisation strategies to support Nvidia Kepler architecture that results in an at least 20 per cent increase in applications’ performance. ttgLabs will also demonstrate how TTG Apptimizer v1.1 can be efficiently applied to video processing in real time.
Xyratex (#364) is a provider of data storage technology, including modular solutions for the enterprise storage industry and hard disk drive (HDD) capital equipment for the HDD industry. Xyratex has more than 25 years of experience in R&D relating to disk drives, storage systems, and manufacturing process technology. The ClusterStor family of HPC storage solutions delivers the ultimate in Lustre performance, reliability, and ease of management.