HP and Voltaire
Hewlett-Packard Company, the market leader in High Performance Technical Computing, offers the broadest spectrum of high performance computing (HPC) solutions. Voltaire InfiniBand switches and HP provide an extremely effective solution for applications demanding high bandwidth and low latency across multiple vertical markets, including systems in the HPC Top 500.
Voltaire (NASDAQ: VOLT) is a leading provider of scale-out computing fabrics for data centers, high performance computing and cloud environments. Voltaire’s family of server and storage fabric switches and advanced management software improve performance of mission-critical applications, increase efficiency and reduce costs through infrastructure consolidation and lower power consumption. Voltaire products are used by more than 30 percent of the Fortune 100 and other premier organizations across many industries, including many of the TOP500 supercomputers.
A member of HP’s Developer & Solution Partner Program (DSPP) since 2004, Voltaire complements the simplicity, agility, and value of the HP Unified Cluster Portfolio; together they provide an extremely effective solution for data centers with demanding HPC applications.
HP Cluster Platforms
Clustering is simple and fast with HP Cluster Platforms, the foundation of the Unified Cluster Portfolio. These systems combine the flexibility of a custom solution with the simplicity, reliability and value of a preconfigured, factory-built product.
HP Unified Cluster Portfolio configurations based on Voltaire InfiniBand switches are tuned for performance and reliability, increasing investment value by lowering total cost ownership and increasing productivity.
The Cluster Platforms and Voltaire solutions provide outstanding scalability to hundreds or thousands of nodes and offer a better choice for HPC environments because it lowers costs, improves quality, and drives breakthrough price/performance.
Success in the Field
Audi was looking for a high-performance computing system to increase safety simulation throughput and helps them speed up design time of vehicle development. After an extensive selection process, Audi chose an HP supercomputer based on Voltaire’s InfiniBand switches and software for the supercomputer's cluster interconnect.
Audi's 29.184-teraflop (peak performance) system is based on HP Cluster Platform 3000BL, a system built using 608 nodes supporting HP ProLiant BL460c server blades connected by Voltaire Grid Director 9024 20 Gigabits/second (Gb/s) double data rate (DDR) InfiniBand switches. The system uses ESI's PAM-CRASH 2G simulation software to perform crash simulations for various Audi car models.
This solution greatly improves the simulation software's performance, allowing for deeper and faster analysis and decreasing the design time needed for new car models.
Voltaire Scale-out InfiniBand
InfiniBand is a high performance, switched fabric interconnect standard for servers. The technology is widely deployed worldwide in server clusters ranging from two to thousands of nodes. InfiniBand has become the standard interconnect of choice for high performance computing (HPC) environments and is becoming the preferred standard in enterprise data centers requiring a high performance fabric.
Today, many high performance data centers take advantage of InfiniBand-based interconnect support incorporated into leading ISV applications and MPI-based applications. Voltaire InfiniBand interconnects eliminate I/O bottlenecks, allowing these HPC applications to run faster and more efficiently.
Voltaire Scale-out Ethernet
Voltaire’s scale-out 10Gb Ethernet offerings enable users to benefit from a far more scalable, lower latency, and virtualized fabric with lower overall fabric costs and power consumption, greater efficiencies, and simplified management than traditional 10 Gb Ethernet fabrics.
Instead of using many hierarchical switching tiers that are constrained by the performance of its root, the Voltaire approach allows greater scalability while guaranteeing higher performance, simpler management, and lower costs. For a customer building a 1000 node data center, this approach delivers 10X lower latency and 4X faster core performance for half the price, using 3X less power than alternative solutions.
Delivering Scale-Out Data Center Fabrics
Voltaire delivers scale-out computing fabrics for data centers, high performance computing and cloud environments. Voltaire’s family of server and storage fabric switches and advanced management software improve performance of mission-critical applications, increase efficiency and reduce costs through infrastructure consolidation and lower power consumption.
Voltaire allows customers to choose the fabric that is right for them – InfiniBand or 10 Gigabit Ethernet – based on their application requirements and individual technology practices, without sacrificing scalability and performance.
Voltaire provides high-performance InfiniBand connectivity and microsecond latencies to enable high-performance applications to run on distributed server, network and storage resources. Applications can gain up to a 300% performance improvement using Voltaire switches as the cluster interconnect instead of Ethernet. Moreover, the switches' power-efficient design offers lower power and cooling requirements as compared to 1 and 10 Gigabit Ethernet offerings.
Voltaire OFED Support
Voltaire offers an easily installable binary form of the Open Fabric Alliance’s Open Fabric Enterprise Distribution (OFED) along with the most recently available fixes and patches that have been discovered during the Voltaire testing and certification process. All such fixes and enhancements to OFED are continually contributed back to the OpenFabrics developer community so they may be used in future versions of OFED.
Supported OFED software components include:
|
|
Voltaire Messaging Accelerator™ (VMA™) Software
Voltaire Messaging Accelerator™ is a software solution that dramatically improves performance of multi-cast applications running on InfiniBand fabrics. Voltaire Messaging Accelerator boosts performance for message based applications such as those found in financial service environments that use Wombat and Reuters RMDS. The result is a reduction in latency by as much as 50% and an increase in application throughput per server as compared to applications running on standard Ethernet interconnect networks.
Voltaire’s Unified Fabric Manager™ (UFM™) Software
Voltaire’s Unified Fabric Manager (UFM) is a powerful platform for managing demanding scale-out computing environments. Now data center operators can efficiently monitor and operate the entire fabric, increase application performance and maximize fabric resource utilization.
UFM uses an innovative application-centric approach to bridge the gap between servers, applications and fabric elements. UFM's revolutionary fabric model allows users to manage fabrics as a set of business related entities such as time critical applications or services. UFM’s management infrastructure enables fabric monitoring and performance optimization on the application-logical level rather than just at the individual port or device level.
Click here for a list of Voltaire HP WW & regional sales contacts.