Online resource kit >

Windows HPC Server 2008 and System Center Server Management Suite Enterprise Better Together

This paper describes the benefits of using a Microsoft System Center Server Management Suite Enterprise (SMSE) solution to deploy, configure, monitor, and back up Windows HPC Server 2008–based clusters. The paper begins with a description of current high-performance computing (HPC) management challenges: inefficiencies in deployment, integration, and management of nodes, lack of agility, and increased cost of ownership. Together, these challenges reduce end-user productivity as HPC users must wait longer for problems to be resolved, for new jobs to be added to the cluster, and for required configuration changes to take place. These challenges also form barriers to HPC usage.

The paper then introduces the benefits of using Microsoft System Center management solutions for HPC platforms. System Center management solutions:

Provide efficient deployment, integration, and management of HPC compute cluster nodes. By automating the deployments of operating systems and application images, a System Center solution reduces “high-touch” manual deployments and speeds up cluster deployments. Robust monitoring and reporting helps reduce system configuration drift, which helps keep cluster nodes optimized.
Make the HPC data center more agile. The deployment, automation, and monitoring features in System Center solutions help organizations utilizing HPC IT investments by scaling to larger more complex HPC clusters. System Center solutions allow for greater cluster utilization and efficiency, thus driving better business results. System Center solutions also help provide a more productive HPC environment that is more responsive to changing requirements, increasing user satisfaction.

Reduce cost of ownership. Microsoft System Center Operations Manager can monitor entire HPC clusters at scale, reducing the number of administrative staff required to manage the HPC environment. Robust alerting reduces the time required to identify and resolve problems, reducing downtime. Since System Center solutions are often used to manage the larger IT environment in many organizations, using a System Center solution to manage the HPC infrastructure utilizes existing skill sets within the organization, reducing the need for specialists to manage the HPC infrastructure and reducing administrative costs.