Source: Platform Computing
Platform Computing, the global leader for grid computing software solutions, today announced the integration of Platform LSF with Microsoft Windows Compute Cluster Server 2003, which allows IT to allocate resources in real time through a single management console, leveraging the performance and capacity of both cluster and data centre environments.
Powered by Platform Enterprise Grid Orchestrator (EGO), Platform LSF is recognised as an industry-leading solution for resource management and orchestration for compute and data intensive applications. Through bi-directional job forwarding enterprise IT benefits from a more adaptive infrastructure that can offload jobs wherever optimal capacity exists – to or from the compute cluster or data centre. The Platform LSF integration for Windows Compute Cluster Server 2003 is scheduled for release in the first half of 2006.
"The market for virtual environment software, the foundation for high performance computing (HPC), is growing rapidly as organisations seek ways to make their IT investments demonstrate measurable benefits in a very short order," said Dan Kusnetzky, Vice President, System Software, Enterprise Computing Group, IDC. "Often the virtualisation technology commercial organisations are evaluating has proven itself supporting other workloads. As commercial organisations grow comfortable with this approach, it becomes the underpinning for many of its future IT investments. The work between Platform Computing and Microsoft will provide exactly the type of technology needed for these distributed applications."
In addition to the announced integration, Microsoft and Platform Computing have agreed to collaborate on the development of an open standard for web services job scheduling.
"Microsoft is focused on removing cost and complexity from high performance computing deployments, and delivering unprecedented computing power to the workgroup and department," said Kyril Faenov, Director, High Performance Computing, Microsoft Corp. "Through work with HPC industry leaders like Platform Computing, we have the ability to extend our solutions across heterogeneous environments and provide integrated workload management – improving overall performance and efficiency for our customers."
"As organisations aggressively embrace low-cost cluster solutions within the HPC and data centre environments there is increased challenge in how workflow is distributed and managed," said Chris Purpura, vice president, New Ventures and Alliances, Platform Computing. "Through a seamless integration with Windows Compute Cluster Server 2003, we will provide both Platform and Microsoft customers with more control, flexibility and agility to respond to the dynamic data processing needs within their HPC and data centre environments."
Also today, Platform Computing announced the integration of Platform LSF with IBM's eServer Blue Gene/L Solution. Powering many of the world's top supercomputers, IBM's Blue Gene/L is optimized for bandwidth, scalability and power efficiency. Through integrating with Platform LSF, powered by Platform Enterprise Grid Orchestrator (EGO), Blue Gene/L now has access to the industry's leading solution for resource management and orchestration of compute and data intensive applications. This integration provides Blue Gene /L users with the ability to concurrently process multiple jobs of variable size - which in turn improves utilization and increases response time. In addition, Platform LSF also extends advanced workload management across heterogeneous platforms. The integration of IBM Blue Gene and Platform LSF is expected to be complete in Q1 2006.
"IBM Blue Gene was designed to deliver new levels of performance and scalability while maintaining impressive power consumption standards," said Herb Schultz, Program Director, Deep Computing Marketing, IBM. "Through working with software providers such as Platform Computing, we can leverage our respective strengths to enhance the overall workload management experience for Blue Gene customers - and can make it easier and more attractive for Platform LSF customers to deploy the Blue Gene solution."
The National Center for Atmospheric Research (NCAR) provides research, facilities, and services for the atmospheric and Earth sciences community.
Currently, NCAR employs IBM Blue Gene to test the viability of the Blue Gene architecture for various geosciences codes including climate simulation software. NCAR uses Platform LSF for managing batch processing on several of its production systems. With the impending integration of Platform LSF and Blue Gene, NCAR will be able to use the same tools to manage both Blue Gene and the rest of their production HPC environment.
"As atmospheric research takes a more prominent role on the world stage, it is increasingly important that we take steps to ensure we have adequate computational resources available for geoscience simulation," said Richard Loft, Deputy Director of Research and Development in the Scientific Computing Division at NCAR. "The impending integration of Platform LSF and IBM Blue Gene will allow us to harness the full potential of our HPC environment and facilitate our ability to manage multiple workloads. These enhancements will give us the ability to more tightly integrate Blue Gene/L with our existing systems."
"Widely, deployed by leading organizations around the world, Platform LSF is recognized as the industry's leading solution for workload management solution for compute and data intensive applications," said Chris Purpura, Vice President, Alliances and New Ventures, Platform Computing. "Through expanding an already strong relationship with IBM, we provide both Platform LSF and Blue Gene customers with an attractive proposition to leverage capacity to the fullest and yield increased benefits from their HPC infrastructure."
Separately, Platform Computing announced that Platform Rocks Standard Edition (SE) will be available as a cluster management solution for the HP Cluster Platform 3000.
Platform Rocks SE is a commercially supported and enhanced implementation of the open source cluster management toolkit, NPACI Rocks. Platform Rocks SE provides customers a standardized tool to simplify management and make it easy and cost-effective for users to rapidly deploy, configure and monitor their cluster environment.
Designed to be flexible, reliable and easy to deploy, HP Cluster Platform 3000 based on Intel Xeon provides a modular, standards-based platform designed to handle the high performance computing (HPC) needs for clusters between 5 and 1024 nodes.
The addition of Platform Rocks SE represents the expansion of a long-standing relationship between Platform and HP in the HPC market. Platform Rocks SE will be available for the HP Cluster Platform 3000 this month.
"As commodity-based Linux clusters grow in popularity, the market is demanding solutions that require minimal IT and human resources to configure and manage," said Chris Purpura, Vice President, New Ventures and Alliances, Platform Computing. "With Platform Rocks SE, we have the ability to enhance the configuration and management process for HP Cluster Platforms, increasing overall performance and efficiency to provide a lower total cost of ownership for the customer."
"HP Cluster Platforms are built on industry-standard technology and offer the widest span of customer choice and flexibility, backed by HP," said Bruce Toal, Director Marketing and Solutions HPC, HP. "The availability of Platform Rocks SE as a management tool for the HP Cluster Platform 3000 expands our customers' choices even more, providing a cost-effective and commercially supported means to deploy the popular NPACI Rocks Linux cluster solution."