XK Series

Cray XK Series Supercomputers

Bringing Cray’s trademark reliability, flexibility and scalability to the many-core HPC environment

Talk to a Cray Expert

CONTACT US NOW

Cray® XK™ Series Supercomputers

The Cray® XK7™ supercomputer combines Cray’s proven Gemini interconnect, AMD's leading multicore scalar processors and NVIDIA’s powerful many-core GPU processors to create a true, productive hybrid supercomputer.

Capable of scaling to 500,000 scalar processors and 50 petaflops of hybrid peak performance, every aspect of the Cray XK7 system — from its resiliency features to its scalability-boosting technologies — has been engineered to meet science’s real-world demands. And with the expertise of the industry’s only company singularly focused on high performance computing behind it, the Cray XK7 supercomputer brings Cray’s trademark reliability, flexibility and scalability to the many-core HPC environment.

Featured Resources

Cray XK7 Brochure

The Cray XK7 supercomputer is a trifecta of scalar, network and many-core innovation.

Four Large-Scale Science Applications Sustain 1+ Petaflops on Cray XE/XK System “Blue Waters”

Working with Cray on optimizations, researchers ran four of the applications (VPIC, PPM, QMPACK and SPECFEM3D_GLOBE) above 1 petaflops of sustained performance.

The Cray XK7 Big Red II: Embracing the spirit of Indiana University

"Big Red II" helps researchers accelerate new breakthroughs in fields that are changing the way we live, work, learn and grow.

See All Resources

Cray® XK7™: Scalable Many-Core Performance

Adaptive Hybrid Computing

The Cray® XK7™ compute node combines AMD's 16-core Opteron™ 6200 Series processor and the NVIDIA® Tesla® K20 GPU Accelerator to create a hybrid unit with the intra-node scalability, power-efficiency of acceleration and flexibility to run applications with either scalar or accelerator components. This compute unit, combined with the excellent inter-node scalability of Cray’s Gemini interconnect, creates a system geared for any computing challenge.

Production Reliability

Integrated Hardware Supervisory System

Cray's Hardware Supervisory System (HSS) integrates hardware and software components to provide system monitoring, fault identification and recovery. An independent system with its own control processors and supervisory network, the HSS monitors and manages all major hardware and software components in the Cray XK7 supercomputer. In addition to providing recovery services in the event of a hardware or software failure, HSS controls power-up, power-down and boot sequences, manages the interconnect, reroutes around failed interconnect links, and displays the machine state to the system administrator. 

Cray XK7 System Resiliency

The Gemini interconnect is designed for large systems in which failures are to be expected and applications must run to successful completion in the presence of errors. Gemini uses error correcting code (ECC) to protect major memories and data paths within the device. In addition, the Cray Linux Environment (CLE) features NodeKARE™ (Node Knowledge and Reconfiguration). If a program terminates abnormally, NodeKARE automatically runs diagnostics on all involved compute nodes and removes any unhealthy ones from the compute pool. Subsequent jobs are allocated only to healthy nodes and run reliably to completion. The XK7 supercomputer’s Lustre® file system can be configured with object storage target failover and metadata server failover. 

Adaptive Supercomputing 

Extreme Scale and Cluster Compatibility

The Cray XK7 system provides complete workload flexibility — a single machine can run both a highly scalable custom workload and industry-standard ISV workload. CLE accomplishes this through Cluster Compatibility Mode (CCM). CCM allows immediate compatibility with Linux/x86 versions of ISV software — without recompilation or relinking — and allows for the use of various versions of MPI (such as, MPICH and Platform MPI™). The service is dynamic and available on an individual job basis.

Support for Other File System and Data Management Services

You can select the Lustre parallel file system or another option including connecting to an existing parallel file system. The Cray Data Virtualization Service allows for the projection of various other file systems (including NFS, GPFS™, Panasas® and StorNext®) to the compute and login nodes on the Cray XK7 system. Cray can also provide solutions for backup, archiving and data lifecycle management.

Power Efficiency

Many-core processing is the key to ultimate energy efficiency. Applications using the Cray XK7 GPU processors will experience industry-leading energy efficiency when measured for real application workloads. Combined with our standard air- or liquid-cooled high-efficiency cabinet and optional ECOphlex™ technology, the Cray XK7 system can reduce cooling costs and increase flexibility in datacenter design.

Software for the Cray® XK7™ System

The Cray® XK7™ supercomputer supports a full range of powerful scalar tools, libraries, compilers, operating system and third-party software. 

Programming Environment

Each Cray XK7 system includes a fully integrated Cray programming environment with tools designed to maximize programmer productivity and application scalability and performance. The XK7 supercomputer can use a variety of high performance Fortran, C and C++ compilers and libraries for both x86 and GPU workloads. Available compilers include GPU-enabled PGI, CAPS, PathScale and the Cray Compiling Environment (CCE) for x86 with support for the ISO Fortran standard (2008) including parallel programming using coarrays, C/C++ and UPC. High performance-optimized math libraries include BLAS, FFTs, LAPACK, ScaLAPACK, SuperLU and Cray Scientific and Math Libraries.

Programming Models

Supported parallel programming models include MPI, CUDA, Cray SHMEM, UPC, Coarray Fortran, OpenMP and OpenACC. The MPI implementation is compliant with the MPI 2.0 standard and is optimized to take advantage of the Gemini interconnect. Cray’s performance and analysis toolkit, CrayPAT™, allows users to analyze resource utilization throughout their code at scale and eliminate bottleneck and load imbalance issues. In addition, the Gemini interconnect has direct hardware support for partitioned global address space (PGAS) programming models including Unified Parallel C (UPC), Co-Array Fortran and Chapel.

Scalable Software

The Cray XK7 system includes the latest Cray Linux Environment (CLE), a suite of high performance software including a SUSE® Linux®-based operating system designed to scale efficiently and run large, complex applications. With highly scalable applications, Compute Node Linux (CNL) runs in Extreme Scalability Mode, ensuring operating system services do not interfere with application scalability. Additionally, the Cray XK7 system provides for tightly integrated, industry-standard batch schedulers, compilers and debuggers.

OpenACC

Cray is a joint founding member of OpenACC, an organization that is driving a new directives-based open parallel programming standard enabling scientific and technical programmers to more easily leverage the performance optimization capabilities of hybrid CPU/GPU computing systems. Learn more by visiting www.OpenACC.org.

Product Brochures

Cray XK7 Brochure

The Cray XK7 supercomputer is a trifecta of scalar, network and many-core innovation.

Case Studies

Researchers Create Largest-Ever Storm Prediction Model to Simulate Hurricane Sandy Using ‘Blue Waters’ — NCSA’s Cray® XE6™ Supercomputer

The devastation caused by Hurricane Sandy and other super storms like it signals the need for further advances in the accuracy and reliability of numerical weather prediction. 

NCSA’s Cray XK7 Supercomputer ‘Blue Waters’ Enables Astrophysics Team to Simulate a Key Supernova Phase at Unprecedented Resolution

Type Ia supernovae (SNe Ia) are the biggest thermonuclear explosions in the modern universe. Because of their brilliance and nearly constant luminosity at peak, these supernovae are also one of the favored “standard candles” (astronomical objects with a known luminosity) used by cosmologists to measure the expansion rate of the universe.

Cray System “Blue Waters” Helps Researchers Develop a Defense Against Damaging Space Weather

The solution — and the challenge — to coping with the effect of solar storms on Earth and its technological systems is forecasting space weather.

Using Cray System “Blue Waters,” Researchers Break the HIV Capsid Code and Open the Door to New Virus-Busting Therapies

Researchers have long sought to understand how the HIV capsid — a protein shell that protects the virus’s genetic material — is constructed.

Four Large-Scale Science Applications Sustain 1+ Petaflops on Cray XE/XK System “Blue Waters”

Working with Cray on optimizations, researchers ran four of the applications (VPIC, PPM, QMPACK and SPECFEM3D_GLOBE) above 1 petaflops of sustained performance.

Titan Simulates Earthquake Physics Necessary for Safer Building Design

Researchers conduct unprecedented study on GPUs of high-frequency shaking

GPUs and the Science of Smooth Ice Cream

Cray XK6 system helps scientists accelerate research into soft matter

Applications for Cray Systems

Tuning NAMD on the Cray XK6 "Titan" Supercomputer

With Cray support, the NAMD developers are optimizing their code on each new iteration of Cray hardware, achieving scaling to hundreds of thousands of cores.

Adaptive Computing: Moab HPC Suite for Cray

With Moab HPC Suite from Adaptive Computing, Cray users get the highest workload productivity and service levels from systems.

Allinea Development Tools for Cray Supercomputers

Leading corporations and research institutions across the world are using Allinea Software products, which set the standard for capability, performance and ease of use.

Altair PBS Professional for Cray

Altair has extensively re-architected PBS Professional for Cray users, collaborating closely with Cray to ensure tight, seamless integration.

ANSYS Fluent for Cray

The high performance computing solution from Cray and ANSYS is a key enabler of high-fidelity simulations, providing a combined solution that facilitates large and detailed aeroelastic simulations.

Rogue Wave Products Engineered for Cray

Developers of applications for Cray platforms can leverage Rogue Wave tools and components to improve productivity and fully harness the power of the hardware.

Videos

The Cray XK7 Big Red II: Embracing the spirit of Indiana University

"Big Red II" helps researchers accelerate new breakthroughs in fields that are changing the way we live, work, learn and grow.

OpenACC Programming for Cray Supercomputers

Parallel programming has never been this easy.

Frontier Scientists: A Cray Supercomputer Called Fish

For more information visit FrontierScientists.com

NVIDIA: Scientists use a Supercomputer to Improve Ice Cream

One of life's simple pleasures is so complicated that it takes a supercomputer to understand it.

Frontier Scientists: What's a FLOP?

Greg Newby, director of the Arctic Region Supercomputing Center at the University of Alaska, Fairbanks, and Per Nyberg, Cray Inc., define flops and update us on the future of supercomputing and computational science.

CUDACast #3 - Your First OpenACC Program

In this CUDACasts video, we'll see how to take an existing application and, using a few OpenACC directives, accelerate the computationally intensive loops on the GPU using the Portland Groupp's pgcc compiler.

OpenACC Basics

"Introduction to OpenACC and CUDA Programming on the Cray XK7 Platform," course organized by the Swiss National Supercomputing Centre (CSCS), Lugano, Switzerland, February 2013

An Introduction to OpenACC

OpenACC and CUDA Programming on the Cray XK7 Platform

Using the Cray Programming Environment

Introduction to OpenACC and CUDA Programming on the Cray XK7 Platform

To view a full list of our partners, click here.

Solution Partners

AMD and Cray

Cray and AMD Supercomputing Solutions

Cray and AMD work together to create innovative supercomputer solutions with best-of-breed performance, scalability and reliability. These supercomputer solutions feature AMD Opteron™ processors delivering up to 16 cores per socket for scaling in thread-intensive environments with AMD-P power management maximizing performance per watt across workloads. With a solution for every budget and requirement, Cray and AMD make it easy to take advantage of the latest advances in high performance computing.

AMD Opteron Solutions >

NVIDIA and Cray

Cray and NVIDIA Supercomputing Solutions

Cray and NVIDIA® are working together to create innovative supercomputer solutions with best-of-breed performance, scalability and reliability. These solutions feature the NVIDIA® Tesla® K40 GPU accelerator offering 2,880 cores, 1.43 teraflops of peak double-precision performance and 12 GB of GPU accelerator memory, making it possible to process datasets twice as large as prior-generation GPUs. The Tesla K40 also outperforms CPUs by up to 10 times and delivers additional performance with its GPU Boost feature, converting power headroom into user-controlled performance boost. With a solution for every budget and requirement, Cray and NVIDIA make it easy to take advantage of the latest advances in high performance computing.

NVIDIA Tesla Solutions >

GPU-Enabled Content Partners

ANSYS, Inc.

ANSYS, Inc., founded in 1970, develops and globally markets engineering simulation software and technologies used by engineers and designers across a broad spectrum of industries. ANSYS technologies help product development organizations create innovative products and processes that meet customer and cost requirements and high-quality products and processes that reliably perform as intended, while shortening time-to-market and reducing engineering costs. Simulation delivers these benefits by enabling an information-based product development process, with virtual prototyping to optimize designs at earlier stages of development.

Website >

EM Photonics

EM Photonics maximizes the performance of applications via algorithm modification, tuning for the HPC environment and the exploitation of GPU co-processing. The company's solvers can be integrated into customers' systems and optimized for their needs. Whether they’re used with CPU, GPU, FPGA or supercomputers, they create novel solutions. EM Photonics provides off-the-shelf tools such as the CULA product line, and develops tailored solutions for customers in areas such as computational fluid dynamics, computational electromagnetics, bio/life sciences, finance and signal processing.

Website >

NVIDIA

NVIDIA, based in Santa Clara, California, awaked the world to the power of computer graphics when it invented the GPU in 1999. Since then, it has consistently set new standards in visual computing with breathtaking, interactive graphics available on devices ranging from tablets and portable media players to notebooks and workstations. NVIDIA's expertise in programmable GPUs has led to breakthroughs in parallel processing which make supercomputing inexpensive and widely accessible. The company holds more than 1,800 patents worldwide, including ones covering designs and insights that are essential to modern computing. Its Tesla product line uses the GPU's parallel processing capabilities to deliver massive performance for scientists and researchers.

Website >

The Portland Group

The Portland Group develops and markets high performance Fortran, C and C++ compilers and tools. Offering world-class performance and comprehensive cross-platform support and optimization, PGI supports 32-bit x86, 64-bit AMD64 and Intel 64 processor-based systems running over 30 popular Linux and Mac OS X and Windows releases. Advantages include fast compilation speed, integrated graphical parallel debugger and profiler, comprehensive documentation, online tutorials and outstanding technical support.

Website >

Other Partners

Accelrys, Inc.

Accelrys develops scientific business intelligence software and solutions for the life sciences, energy, chemicals, aerospace and consumer products industries. Its customers include many Fortune 500 companies and other commercial entities, as well as academic and government entities. Accelrys has a vast portfolio of computer-aided design modeling and simulation offerings that help customers conduct scientific experiments ”in silico” in order to reduce the duration and cost of discovering and developing new drugs and materials.

Website >

Adaptive Computing

Adaptive Computing provides intelligent automation software for datacenter, cloud and HPC environments. The company’s infrastructure intelligence solutions, powered by Moab®, deliver policy-based governance, allowing customers to consolidate and virtualize resources, allocate and manage applications, optimize service levels and reduce operational costs. Adaptive Computing products manage the world’s largest computing installations and are the preferred intelligent automation solutions for the leading global datacenter vendors.

Website >

Allinea Software

Allinea Software is the trusted leader in software development tools for parallel programming and high performance computing (HPC). Leading users of HPC turn to Allinea for extremely scalable, capable and intuitive tools that improve the efficiency and value of their HPC investment by reducing development time and increasing application performance. Allinea’s integrated suite of profiling and debugging tools is relied on from desktops through to beyond petascale for applications ranging from climate modeling to astrophysics, and from computational finance to engine design.

Website >

Altair Engineering, Inc.

Altair Engineering, Inc., empowers client innovation and decision-making through technology that optimizes the analysis, management and visualization of business and engineering information. Altair is the only company to create HPC solvers and simulation applications, deliver HPC management tools, and use these applications and tools to solve real HPC issues. With a track record for high-end software and consulting services for engineering, computing and enterprise analytics, Altair consistently delivers a competitive advantage to customers in a broad range of industries.

Website >

CD-adapco

CD-adapco is the leading global provider of full-spectrum engineering simulation solutions for fluid flow, heat transfer and stress, including CFD software, CAD-embedded CFD, CAE consulting services and training. STAR-CD and STAR-CCM+ provide the world’s most comprehensive CFD solutions. Their integrated CAE process allows engineers to focus more on product design by giving them a single easy-to-use interface to automatically handle complex models through the complete analysis cycle: preprocessing, meshing, solution, post-processing. CD-adapco’s solutions are used across many industries, including automotive, aerospace, appliances, buildings, chemical process, electronics, environmental, marine, offshore structures, pharmaceuticals, power generation, rail, turbomachinery and specialized mechanical engineering applications.

Website >

CEI

CEI delivers the world's most advanced visualization technology for large, complex data. CEI was spun out of Cray in 1994. CEI's EnSight family of products provides a comprehensive visualization and post-processing suite to help understand scientific and engineering simulation data. CEI has offices in Detroit, Houston, Raleigh, Munich, and Shanghai, and distributors in other regions around the world.

Website >

Livermore Software Technology Corporation (LSTC)

Livermore Software Technology Corporation develops LS-DYNA® and related engineering software products. LS-DYNA is a general-purpose transient dynamic finite element program that simulates complex real-world problems. It includes a Pre/Post Processor, LS-PrePost®, and a tightly integrated optimization code, LS-OPT®. LS-DYNA, LS-PrePost and LS-OPT are used in various industries, including automobile design, aerospace, manufacturing and bioengineering. LS-DYNA development is focused on one code methodology that includes implicit, explicit, SMP and MPP solvers. It is optimized on all platforms including clusters running Unix, Linux and Windows.

Website >

MathWorks

MathWorks is the leading developer of mathematical computing software. MATLAB, the language of technical computing, is a programming environment for algorithm development, data analysis, visualization and numeric computation. Simulink is a graphical environment for simulation and model-based design of multidomain dynamic and embedded systems. Engineers and scientists worldwide rely on these product families in the automotive, aerospace, electronics, financial services, biotech-pharmaceutical and other industries. MathWorks products are also used in universities and learning institutions worldwide.

Website >

Metacomp Technologies, Inc.

Metacomp Technologies is at the forefront of technology in computational fluid dynamics, computational aero-acoustics, structural analysis and mesh generation. Metacomp is provides  software tools for aerospace, defense and automobile manufacturers, U.S. government labs and research organizations, as well as other companies that require complex modeling during research and development. Its activities include R&D, software development, dissemination and support in the areas of fluid dynamics, structural analysis, numerical algorithms, turbulence and combustion modeling, acoustics, geometry modeling and mesh generation.

Website >

MSC Software

MSC Software is the worldwide leader in multidiscipline simulation. MSC Software helps companies improve quality, save time and reduce costs associated with design and testing of manufactured products. Leading manufacturers use its engineering simulation technology for linear and nonlinear finite element analysis (FEA), acoustics, CFD, multi-physics, optimization, fatigue and durability, multi-body dynamics and control systems simulation. The company’s products accurately and reliably predict how products will behave in the real world to help engineers design more innovative products, quickly and cost-effectively.

Website >

Numerical Algorithms Group

The Numerical Algorithms Group (NAG) delivers high-quality computational software and high performance computing services. For 40 years NAG experts have worked closely with world-leading researchers in academia and industry to create powerful, reliable and flexible software that is relied on by tens of thousands of individual users, as well as numerous independent software vendors. NAG serves its customers from offices in Oxford (U.K.), Chicago, Tokyo and Taipei, through staff in France and Germany, and via a global network of distributors.

Website >

Platform Computing

Platform Computing is a pioneer and the global leader in high performance computing (HPC) management software. The company delivers integrated software solutions that enable organizations to improve time-to-results and reduce computing costs. Many of the world’s largest companies rely on Platform to accelerate compute- or data-intensive applications and manage cluster and grid systems. Platform has over 2,000 global customers and strategic relationships along with the industry’s broadest support for HPC applications.

Website >

Rogue Wave Software

Rogue Wave is the largest independent provider of cross-platform, embeddable libraries and development tools for HPC application design, testing and deployment. Rogue Wave’s tools and components empower software developers to take advantage of the performance and scalability that multicore and parallel technologies offer, and are proven by decades of use solving some of the world’s most complex problems in commercial enterprise, academia, defense and institutional research.

Website >

SIMULIA

SIMULIA provides a scalable portfolio of simulation solutions and an open platform to support integration of multidisciplinary analysis with its partners. Simulation enables engineers and scientists to improve product performance, eliminate physical prototypes and drive innovation. The product suite includes the Abaqus/CAE® modeling, visualization and automation environment for ABAQUS analyses; Abaqus/Standard® for static, dynamic, thermal and coupled analysis with implicit FEA; Abaqus/Explicit® for transient dynamic and quasi-static analysis with explicit FEA; and Abaqus for CATIA V5® for advanced nonlinear analysis within CATIA V5.

Website >

Software Cradle Co., Ltd.

Established in 1984 and headquartered in Osaka, Japan, Software Cradle Co., Ltd., develops and markets innovative computational fluid dynamics software for a broad range of customers. Software Cradle Co. provides customers with leading-edge technology that can open new frontiers in the design of their products.

Website >

Tech-X Corporation

Tech-X was founded in 1994 to develop products and services for leveraging high performance computing in computational physics. Since then, Tech-X has provided services to a broad range of national and international customers from government, academia and industry. Tech-X Corporation is committed to technical excellence and innovation. Its core competencies include the simulation of processes, devices and physical systems related to plasma physics, fusion and accelerator technologies, performed on computer systems ranging from desktops to leadership computing facilities.

Website >