Cluster management
An Extreme Computing cluster needs to have all the qualities you would expect from an effective production system: it must be available and capable of delivering the best performance under normal working conditions, with a varied and fluctuating workload, multiple users and multiple applications. This means the choice of cluster management suite is crucial, and has a real impact on the effective availability rate of the computing environment. At the same time, users should be able to concentrate on their own applications, rather than spend their time on administering the cluster.
Bull Extreme Computing solutions offer a choice between the bullx cluster suite - Bull’s comprehensive Extreme Computing software suite for Linux environments, formerly known as Bull Advanced Server for Xeon - and Microsoft® Windows® Server 2008.
Linux 
Windows HPCS
Linux 
Bull has designed the bullx cluster suite to make your Extreme Computing cluster as easy to operate and as reliable as a single system.
The bullx cluster suite is a robust and efficient solution that delivers total cluster management. It addresses each step of the cluster lifecycle: installation and software deployment, monitoring and fault handling, cluster optimization and expansion, and provides a comprehensive development environment.
Integrated, tested and supported by Bull
The bullx cluster suite federates the very best of Open Source components, complemented by leading software packages from well-known Independent Software Vendors, and gives them a consistent view of the whole Extreme Computing cluster through a common database. All components have been fully tested and integrated within the bullx cluster suite, and are naturally supported by Bull.
The bullx cluster suite includes:
- A stock RedHat Enterprise Linux 5 distribution, to benefit from the support and certification of virtually any HPC ISV application available for RedHat environments
- Bull System Manager HPC Edition for cluster management and monitoring provides cluster administrators with powerful tools through a single and consistent web-based interface that accelerates their daily operations:
- Software installations and upgrades: deployment over up to 500 nodes in 20 minutes
- Detailed reporting of faulty components: Nagios will display the drawer where a faulty component is located, down to the label of the cable connected to it
- Topological and node group views to focus on specific subsets of your cluster.
- Specific extensions developed and supported by Bull, to meet the specific needs of HPC applications, such as a global shared file system (Lustre) or cutting-edge InfiniBand stack. They are integrated in such a way that the environment remains wholly compatible with the standard RedHat environment.
bullx cluster suite components
|
|
|
Microsoft® Windows® HPC Server 2008
Bull, as a Microsoft Gold Partner, supports Microsoft Windows HPC Server 2008. The bullx clusters can be delivered with Windows HPCS pre-installed, to provide an easy to use, scalable HPC infrastructure.
Microsoft Windows HPC Server 2008 (HPCS), the next generation of high performance computing, provides enterprise-class tools, performance, and scalability for a highly productive HPC environment. HPCS provides a complete and integrated cluster environment including the operating system, a Job Scheduler, Message Passing Interface v2 (MPI2) support, and cluster management and monitoring components. Built on Windows Server® 2008 64-bit technology, HPCS can efficiently scale to thousands of processing cores and includes a management console that helps proactively monitor and maintain system health and stability. Job scheduling interoperability and flexibility enables integration between Windows and Linux-based HPC platforms, and supports batch and service-oriented architecture (SOA) workloads. Enhanced productivity, scalable performance, and ease of use are some of the features that make Windows HPC Server 2008 best-of-breed for Windows environments.
Windows HPC Server 2008 can help shorten user time-to-insight for HPC workloads through easier deployment and management. By using the existing Windows-based information technology (IT) infrastructure, HPCS simplifies management, security, and storage for the cluster, and provides seamless access from the desktop.
HPCS includes improved provisioning based on the Windows Server 2008 Windows Deployment Services technology, a faster Microsoft Message Passing Interface (MS-MPI) that includes new NetworkDirect support, an advanced Job Scheduler, and a new management interface built on the Microsoft® System Center 2007 user interface (UI), which has support for Windows PowerShell™ as a preferred scripting interface.
HPCS uses the failover clustering capabilities of Windows Server 2008 Enterprise and of Microsoft® SQL Server® to provide high availability and redundancy of the head node in your HPC cluster.
| |
|
