United States Change | All Microsoft Sites
Microsoft Home | Servers and Tools
- Home
- Product Info
- Solutions
- Try It
- How to Buy
- Technical Resources
- Community
- Partners
New to HPC?
get acquainted.
A Competitive Advantage
High Performance Computing gives analysts, engineers and scientists the computation resources they need to make better decisions, fuel product innovation, speed research and development, and accelerate time to market. Some examples of HPC usage include: decoding genomes, animating movies, analyzing financial risks, streamlining crash test simulations, modeling global climate solutions and other highly complex problems.
More Accessible Than Ever
In the past, the most common way to apply multiple compute cycles to a complex problem was to use specialized supercomputing hardware – a solution with a very high cost of entry and technical complexity.
However, recent software and hardware advances have made it possible to leverage existing IT skills and create an HPC environment using off-the-shelf servers and high speed interconnects. These systems can deliver industry-leading computing power with more efficiency and at a significantly lower cost of entry and ownership. This form of HPC is called a commodity HPC cluster.
Basic Architecture of an HPC Cluster
A cluster consists of several servers networked together where each server in the cluster performs one or more specific tasks. Cluster components include Head Nodes, and Compute Nodes, Job Scheduler and Broker Nodes (for SOA enabled clusters.)
Head Node
The single point of management and job scheduling for the cluster. It provides failover and controls and mediates access to the cluster resources.
Compute Node
Carries out the computational tasks assigned to it by the job scheduler.
Job Scheduler
Queues jobs and their associated tasks. It allocates resources to these jobs, initiates the tasks on the compute nodes; and monitors the status of jobs, tasks, and compute nodes.
Broker Node
Act as intermediaries between the application and the services. The broker load-balances the service requests to the services, and finally return results to the application.
Learn more: read Deployment Roadmap for Windows HPC Server 2008 R2 »
Basic cluster: The minimal set of
components to run an HPC application on a cluster.
Resilient cluster: Includes redundant Head and Broker nodes so that applications are immune to failures of these critical servers.
Workstation cluster: Includes Windows 7 workstations that can be used as compute nodes to run HPC jobs when they are otherwise idle.
SOA Enabled cluster: Building on the basic cluster, it includes one or more broker nodes required to run SOA-based HPC applications
Azure Cloud cluster: Enables Windows HPC Server jobs to run, partially or completely, in Windows Azure
Azure Cloud cluster: Enables Windows HPC Server jobs to run, partially or completely, in Windows Azure
HPC in Action
See how organizations in various industries are using HPC to gain a competitive advantage or speed research and discovery.
-
Financial Services: Reduced Modeling Time
Financial Services: Reduced Modeling Time
Western & Southern Financial Group chose an actuarial solution from Milliman that runs on Microsoft® high-performance computing (HPC) technology that made it easy for the company’s IT staff to manage and adjust to meet business needs. The company is using their new solution to reduce modeling time by up to 99 percent and decrease their time-to-market for new products.
Life Sciences: Advancing Cancer Research
Life Sciences: Advancing Cancer Research
The Research Informatics Core at Nationwide Children’s Hospital increased the speed at which it could digitize pathology slides and support researchers through faster data computation.
Digital Content Creation: Faster Image Rendering
Digital Content Creation: Faster Image Rendering
Glukoza Production was able to render animations for their high-definition, full-length film 25 percent faster, making it possible for them to save $200,000 in production costs and release their movie four months earlier.
Government: Driving Productivity
Government: Driving Productivity
The Institute of Geology, China Earthquake Administration (IGCEA) deployed Windows® HPC Server 2008 on a 128-node cluster with a dual boot of the operating systems. As a result, they simplified their HPC cluster management, decreased cluster deployment time by 60 percent, and increased IT productivity by 50 percent.
Manufacturing: Breaking Barriers
Manufacturing: Breaking Barriers
Seer Technology not only tackled a gas chromatography problem that industry experts said was impossible, but they also saved $7 million and cut their time to market by 83 percent with HPC.
Higher Education & Research: Speed Research, Extend Software Resources
Higher Education & Research: Speed Research, Extend Software Resources
Today, researchers at Cornell University and other institutions are more productive, because they can focus on their research rather than on the underlying technology. Using an HPC system has given them access to a broader range of commercial and custom applications.
Stork Thermeq Expands Research Capabilities
Stork Thermeq Expands Research Capabilities
Due to the accelerated performance, Stork Thermeq researchers can run more compute jobs and gain deeper insight for improved products and services.
180-Day Free Trial
Install and use Microsoft Windows HPC Server in your existing environment
Download the trial software »
A Flexible Platform for Powerful Apps
Discover the variety of applications that run on Windows HPC Server.
View all applications »
Get Started
See what you need to deploy Windows HPC Server.
View System Requirements »
- © 2010 Microsoft
<
p>Source: What Is Microsoft Windows HPC Server?
