Software
Independent Software Vendors (ISVs)
Cray has invested considerable time and resources to ensure a vibrant ISV ecosystem. Through this dedication to ISVs, our customers can rest assured that their applications run on the Cray CX1 supercomputer.
ISV Solution Partners >>
Intel® Cluster Ready
The Cray CX1 supercomputer is a certified Intel Cluster Ready system. The certification program was designed to establish a common specification among original equipment manufacturers, independent software vendors (ISVs) and others for designing, programming and deploying clusters built with Intel components. For users, the certification means that the Cray CX1 system will support a wide range of Intel Cluster Ready ISV applications out of the box.
Learn more about the Cray CX1 and Intel Cluster Ready >>
HPC Operating Environments
To bring HPC capabilities to desktop-bound engineers and scientists, the Cray CX1 supercomputer offers multiple operating environments:
Windows HPC Server 2008
Cray and Microsoft have partnered extensively to create a superior out-of-the-box experience for the HPC user. The Cray CX1 system ships pre-configured with Windows HPC Server 2008 along with installation routines that make configuration and setup a simple and straightforward process.
The Cray CX1 supercomputer with Windows HPC Server 2008 is an integrated solution stack that provides a simplified, common operating environment. This powerful combination offers consistent and integrated deployment, job submission, status and progress monitoring while delivering maximum compute performance and scalability, from workgroup HPC systems to some of the world's largest supercomputers, to the end user. Additionally, this scalable solution tightly integrates with existing desktop Windows infrastructures, allowing users to extend desktop technology and skills to the realm of HPC computing.
Windows HPC Server 2008 integrates with Windows Active Directory, security technologies and management and deployment tools. It supports traditional HPC MPI applications, as well as new interactive and service-oriented parallel applications using Microsoft .NET platform. In addition, Windows HPC Server 2008 easily integrates with existing Windows and Unix environments, making it an ideal platform for developing and running advanced HPC applications, including support for a high performance networking stack, C/C++, MPI, MPI.Net, F#, C#, Fortran, advanced debugging and tracing tools, including Vampir, along with Windows Performance Toolkit.
Learn more about Windows HPC Server 2008 >>
Red Hat Enterprise Linux
Cray packages Red Hat Enterprise Linux (RHEL) with Clustercorp's Rocks+ to create a complete software environment for Linux users.
Red Hat Enterprise Linux (RHEL) is designed for enterprise deployments providing a cost-effective, flexible and scalable environment. It features support for shared cache and multi-core systems, kernel SMP locking enhancements for improved scalability and IPv4/IPv6 performance enhancements. RHEL supports very large high performance file systems with Global File System (GFS) and support for NFSv4. It is also interoperable with Microsoft Windows file and print (CIFS) systems through Samba. Finally, RHEL provides excellent compiler and debugging capabilities with GCC 4.1 and Glibc 2.4.
Rocks+ is a complete Linux-based HPC software stack on a single DVD, which includes the operating system , cluster management middleware, libraries, compilers, and workload management tools; with enterprise class commercial support from Clustercorp. Rocks+Rolls simplifies cluster management by tightly integrating 3rd party software into Rocks, including the Grid Engine Roll, Torque Roll, OFED Roll, Intel Developer Roll, PGI Roll, TotalView Roll, and more. Each component is installed by simple "checkbox" decisions and managed at a single point of entry.
Platform Cluster Manager
Platform Cluster Manager is a "re-think" of how open clusters are deployed and managed. It is a complete, certified solution that includes all the tools required to quickly deploy, run and manage HPC clusters with unprecedented ease. Platform Cluster Manager is built for optimal performance and uptime, while offering management, support and update benefits that are simply not possible with other cluster management solutions. Benefits include:
- Quick and easy deployment
- Simple cluster management through an intuitive web management interface
- Optimal uptime and productivity – updates applied with no re-booting or re-installation
- Quickly and easily build high performance applications through a fully integrated commercial MPI
- Accelerated application performance with integrated workload management
- Easily scales as your business needs grow
- Fully supported by Cray
Clusters Made Simple with the Cray CX1 and Platform Cluster Manager
Platform HPC Workgroup Manager
Platform HPC Workgroup Manager is the industry's most advanced and complete cluster management solution. It is a combination of a simple-to-deploy cluster management solution, Platform Cluster Manager; a sophisticated workload scheduler, Platform LSF Workgroup Edition; and the market's fastest MPI implementation, Platform MPI. Benefits include:
- Easy to deploy, run and manage
- Maximum utilization and optimal allocation of cluster assets
- Accelerates and scales your applications
- Updates compute nodes without re-installing
- Changes node personalities with ease
- Takes advantage of leading interconnect technologies
Platform HPC Workgroup Manager
ScaleMP
ScaleMP™ is the leader in virtualization for high-end computing, providing higher performance and lower Total Cost of Ownership (TCO). The innovative Versatile SMP™ (vSMP) architecture aggregates multiple x86 systems into a single virtual x86 system, delivering an industry-standard, high-end symmetric multiprocessor (SMP) computer. Using software to replace custom hardware and components, ScaleMP offers a new, revolutionary computing paradigm.
vSMP Foundation enables the creation of a single virtual system by aggregating multiple industry-standard x86 systems. The Cray CX1 and vSMP foundation supports SMP functionality from two nodes (16 cores) up to a maximum configuration across two chassis (128 cores) with 1TB of shared systems memory.
vSMP Foundation uses blade or rack-mounted systems and InfiniBand to offer tremendous price/performance advantages for the HPC market. In essence, it provides a unique way to leverage entry-level systems to reduce the total cost of ownership (TCO). It delivers the operational simplicity of traditional shared-memory systems while keeping the acquisition cost associated with clusters. It is ideal for HPC applications benefitting from large memory, high core-count coupled with shared memory. It also simplifies cluster deployments by reducing the number of operating systems to one.
Cray Inc. is an authorized global reseller of vSMP Foundation.
|