Profile: Cray Inc (CRAY.O)
6 Dec 2013
Cray Inc., incorporated on December 7, 1987, designs, develops, manufactures, markets and services high-performance computing (HPC), systems, including categories of systems commonly known as supercomputers and/or clusters, and provide storage solutions, software and engineering services related to HPC systems to its customers, which include government agencies and government-funded entities, academic institutions and commercial entities. The Company focuses on building product solutions for the Company’s customers in two markets: the supercomputing portions of HPC and Big Data, including storage and data analytics. The Company also provides a range of service offerings around these products that leverage its high quality support and intimate understanding of its customer uses. The Company provides solutions ranging from specific components to integrated systems, focusing on custom-designed hardware, software, packaging, power and cooling solutions.
Cray Supercomputing Systems
The Cray XC30 supercomputer is the Company’s integrated supercomputing product, which delivers on its commitment to an Adaptive Supercomputing architecture providing both extreme scale and sustained performance. The Cray XC30 system provides the HPC user community the advantage of the computational resources of our supercomputers powered by Intel Xeon E5 processors combined with its Aries interconnect, flexible Dragonfly network topology, its powerful and integrated software environment power and cooling technologies.
The Cray XK7 supercomputer combines the proven Gemini interconnect AMD's multi-core scalar processors and NVIDIA's many-core GPUs to create a true, productive hybrid supercomputer. The Cray XK7 supercomputer is capable of scaling to 500,000 processors and more than 30 petaflops of hybrid peak performance. The Cray XK7 supercomputer brings the Company's reliability, flexibility and scalability to the many-core GPU HPC environment. The Company's Cray XK7m supercomputer is designed for the technical enterprise market. It leverages all the advantages of the Cray XE6m systems addressing technical enterprise applications and price points while additionally integrating NVIDIA Tesla GPU accelerators.
The Cray XE6 system supports high density processor configurations of 192 AMD Opteron processor sockets or up to 3,072 processor cores and delivering more than 30 teraflops (30 trillion floating point operations per second) of computational capacity in each cabinet, with system peak and sustained performance designed to exceed ten petaflops. Customers can upgrade to the Cray XE6 system from the Cray XT4, Cray XT5 or Cray XT6 systems by upgrading the network, processors, memory and a compute blade, thereby leveraging their investment over a longer period of time. The Cray XE6 Linux-based operating system efficiently supports the extreme levels of scaling featured in each of the Company's supercomputers as well as a large range of industry applications with its Cluster Compatibility Mode software environment. The Cray XE6 system can be liquid cooled through use of Cray ECOphlex technology or air cooled.
The Cray Xtreme-X supercomputer cluster system offers an energy-efficient, air-cooled architecture featuring high performance, high availability computing. It includes flexible configuration options for a range of data center cooling architecture requirements through the use of air or chilled cooling rear door heat exchangers. The Cray Xtreme-X system is integrated with the HPC Software Stack, software tools compatible with most open source and commercial compilers, tools, schedules and libraries to run complex applications. This solution is also integrated with the Advanced Cluster Engine.
The Cray Xtreme-Cool cluster supercomputer system offers the features and benefits of the Cray Xtreme-X system with energy savings, lower total cost of operation and faster return on investment by requiring fewer or no air conditioning units in the data center. Its design uses warm water liquid-cooling heat exchangers with no chillers, reducing typical energy consumption used to cool the data center by 50%. This system offers high performance and three times more energy efficiency per rack versus traditional air-cooled designs. It also produces 80% heat capture to the warm water for possible heat reuse.
The Company's YarcData business focuses on providing solutions to the Big Data market by extending the Company's supercomputing platform to graph analytics problems. The YarcData uRiKA graph appliance solution includes scalable massively multithreaded processors with a massive shared memory architecture that is ideally suited for tasks such as research discovery, pattern matching, complex searches, scenario development, behavioral prediction, anomaly identification and graph analysis. This system is purpose-built for parallel applications that are dynamically changing, require random access to a shared memory and do not run well on conventional systems. This system is ideal for massive unstructured and irregular data mining problems. The design is based on a Cray compute infrastructure but utilizes custom Cray Threadstorm processors developed for massively multithreaded processing.
Storage and Data Management
The Company's storage and data management division offers storage and data management solutions for HPC and Big Data. The Company is able to rapidly deploy scalable file systems that integrate effectively with computing solutions ranging from third-party Linux clusters to highly integrated Cray supercomputers. The Company's storage systems business offers the choice of two product families for customers. The Company's flagship storage product line, Cray Sonexion 1600, embeds the Lustre parallel file system and other software in an optimal configuration to reduce deployment time, increase reliability and precisely scale performance and capacity. Cray Sonexion offers an optimal combination of performance and capacity, scaling capacity from terabytes to petabytes and with performance from three gigabytes to over one terabyte in a single file system.
The Company's component-based solutions provide a single point of support and ensure customers have access to storage and data management technologies. The Company's component-based solutions utilize software and services provided through the Company's data management software and services suite of offerings. The Company provides a suite of data management and storage connectivity products and services enabling data movement for and management of both its and third-party storage platforms.
Engineering and Customer Services
The Company's Custom Engineering business leverages the Company's amassed intellectual property and technology portfolio, deep domain expertise and HPC know-how to design and build solutions and services designed to match a customer's specific needs. The need for a solution often stems from special processing needs that are often performance, application or capacity related; special environmental needs that might include special size dimension, weight, power and cooling limitations; or interface or integration requirements. The Company provides solutions ranging from specific components to complete integrated systems, focusing on custom-designed hardware, software, packaging, power and cooling solutions to address an HPC customer's challenges in processing, application performance, environmental limitations or integration with distinct equipment.
Suite 1000, 901 Fifth Avenue
SEATTLE WA 98164-2058