Subscribe: Dell Power Solutions
http://www.dell.com/downloads/global/rss/high_performance_computing.xml
Added By: Feedage Forager Feedage Grade B rated
Language: English
Tags:
cluster  clusters  computing hpc  computing  dell  high performance  high  hpc clusters  hpc  performance computing  performance 
Rate this Feed
Rate this feedRate this feedRate this feedRate this feedRate this feed
Rate this feed 1 starRate this feed 2 starRate this feed 3 starRate this feed 4 starRate this feed 5 star

Comments (0)

Feed Details and Statistics Feed Statistics
Preview: Dell Power Solutions

Dell Power Solutions



Dell Power Solutions is your one-stop shop in today's complex IT environment. It offers a forum to help you discover solutions that work, learn best practices from peers, review customer success stories, and examine real-life deployments in the emerging a



Published: Tue,22 August 2006 17:00:00 CST

Last Build Date: Tue,22 August 2006 17:00:00 CST

Copyright: Copyright 2006 Dell Inc. All rights reserved.
 



Running Microsoft Windows HPC Server 2008 on Dell PowerEdge Clusters

Fri, 14 Nov 2008 05:00:00 CST

The Microsoft® Windows® HPC Server 2008 high-performance computing (HPC) platform is designed to simplify the deployment, configuration, and management of HPC clusters while also integrating multiple performance enhancements. Running this platform on clusters of Dell™ PowerEdge™ servers can provide a highly available, high-performance foundation for HPC applications.



Deploying HPC Clusters Using Platform Open Cluster Stack – Dell Edition

Fri, 14 Nov 2008 05:00:00 CST

As high-performance computing (HPC) environments continue to grow in size and complexity, cluster deployment and management have become increasingly challenging. Platform™ Open Cluster Stack – Dell Edition helps address these challenges with an open source, modular, hybrid architecture designed to easily integrate third-party software and help simplify deployment and management for cluster administrators.



Accelerated HPC Productivity with Intel Cluster Ready Solutions

Fri, 14 Nov 2008 05:00:00 CST

The Intel® Cluster Ready program provides a standardized, replicable way to build and run high-performance computing (HPC) clusters, helping simplify cluster deployment and management. By using Intel Cluster Ready–certified Dell™ HPC clusters, organizations can quickly install and configure clusters to begin running registered HPC applications.



The Red Hat HPC Solution: Simplified High-Performance Linux Clusters

Fri, 14 Nov 2008 05:00:00 CST

High-performance computing (HPC) cluster deployments can be complicated, time-consuming, and costly. The Red Hat® HPC Solution enables organizations to deploy enhanced systems quickly and run them efficiently, helping reduce the cost and complexity of their Linux® cluster deployments.



Evaluating Third-Generation AMD Opteron Processors for HPC Workloads

Fri, 14 Nov 2008 05:00:00 CST

Third-generation AMD Opteron™ processors are designed to optimize multi-threaded application performance, and include multiple architectural enhancements over second-generation AMD Opteron processors. To help organizations understand the performance increases possible when upgrading to third-generation processors, Dell engineers performed benchmark tests against a variety of high-performance computing (HPC) workloads.



Evaluating Mellanox ConnectX InfiniBand Performance in HPC Clusters

Fri, 14 Nov 2008 05:00:00 CST

IT organizations need an efficient communication subsystem to sustain application performance. The Dell team tested bandwidth and benchmarks to determine speedup rates in high-performance computing (HPC) clusters used on Dell PowerEdge™ servers. Next-generation InfiniBand ConnectX technology helps address some I/O challenges by improving scalability and performance across symmetric multiprocessing systems.



Building One of the World’s Largest Windows-Based HPC Clusters

Tue, 12 Aug 2008 05:00:00 CST

When the University of Nebraska at Omaha’s Holland Computing Center teamed up with Dell to build a high-performance computing (HPC) cluster, the result was a flexible dual-boot system of Dell™ PowerEdge™ servers with dual-core AMD Opteron™ processors—and one of the world’s largest and most powerful Microsoft® Windows® OS–based clusters.



Creating a Hyper-Efficient Hyper-Scale Data Center

Tues, 5 Feb 2008 05:00:00 CST

A growing number of organizations today utilize computing “clouds” to deliver many applications familiar to Internet users as well as flexible access to powerful compute and storage resources. Integrated solutions designed to minimize acquisition and operating costs, maximize energy efficiency, and enable rapid scalability can help operators of these and other hyper-scale data centers speed deployment, lower total cost of ownership, and reduce environmental impact.



Microsoft HPC Solutions Boost Manufacturing Innovation

Tues, 5 Feb 2008 05:00:00 CST

In manufacturing concerns, high-performance computing (HPC) has evolved from a cost center to a strategic asset. The Microsoft® HPC solution stack facilitates seamless access to distributed computing resourcesenhancing end-user productivity and performance to accelerate global product innovation cycles.



Building Highly Available HPC Clusters with IBRIX Fusion and the Dell PowerVault MD3000

Tue, 13 Nov 2007 05:00:00 CST

Qualified high-performance computing (HPC) cluster configurations based on the IBRIX Fusion file system and Dell™ PowerVault™ MD3000 storage can provide a robust, cost-effective, high-performance architecture for parallel application deployments.



Utilizing Open MPI Middleware for HPC Clusters

Thu, 01 Nov 2007 05:00:00 CST

Open MPI, an open source middleware implementation of Message Passing Interface (MPI), helps simplify the porting and execution of MPI-based applications and supports many high-performance computing (HPC) platforms, interconnects, and environments. This article outlines the Open MPI architecture and its flexible runtime configuration options, and offers tips that can help maximize bandwidth performance.



Parallel Rendering Technologies for HPC Clusters

Thu, 01 Nov 2007 05:00:00 CST

Using parallel rendering technologies with clusters of high-performance computing (HPC) workstations configured with high-end graphics processors helps scale out graphics capabilities by exploiting and coordinating distributed computing resources. This article discusses parallel rendering architectures and highlights open source utilities that can help meet rendering requirements for large-scale data sets.



Intel Cluster Ready and Platform Open Cluster Stack: Clusters Made Simple

Thu, 01 Nov 2007 05:00:00 CST

The Intel® Cluster Ready program is designed to provide a common standard for high-performance computing (HPC) clusters, helping organizations design and build seamless, compatible configurations. Integrating the standards and tools provided by this program with Platform™ Open Cluster Stack and certified Dell™ clusters can help significantly simplify the deployment and management of HPC clusters.



Evaluating the Myrinet-10G Interconnect on Dell PowerEdge Servers

Thu, 01 Nov 2007 05:00:00 CST

The Myricom Myrinet-10G (Myri-10G) interconnect can offer high bandwidth and low latency in high-performance computing environments. This article describes the Myri-10G interconnect architecture and software stack as well as test results comparing the latency, throughput, and performance of Myri-10G, Myrinet-2000, and Gigabit Ethernet on ninth-generation Dell™ PowerEdge™ servers.



Multi-core Scalability with FLUENT and Microsoft Windows Compute Cluster Server 2003

Thu, 01 Nov 2007 05:00:00 CST

Using the FLUENT computational fluid dynamics application in conjunction with Microsoft® Windows® Compute Cluster Server 2003 enables cluster users to easily run large simulations on high-performance computing clusters. This article describes tests performed to evaluate the scalability of these types of clusters when using multi-core processing to help increase performance.



Configuring InfiniBand on HPC Clusters Using the OpenFabrics Enterprise Distribution

Thu, 01 Nov 2007 05:00:00 CST

The OpenFabrics Enterprise Distribution (OFED) provides a standardized software stack for high-performance computing (HPC) clusters using the InfiniBand interconnect. This article describes the development and architecture of OFED along with test results measuring the bandwidth and latency of OFED-based HPC clusters of Dell™ PowerEdge™ servers.



An Introduction to DDR InfiniBan

Wed, 08 Aug 2007 05:00:00 CST

The increasingly popular InfiniBand high-speed interconnect now supports both single data rate (SDR) and double data rate (DDR) transmission, with DDR providing significantly increased speeds. This article discusses how InfiniBand has evolved and presents test results comparing the performance of Gigabit Ethernet, SDR InfiniBand, and DDR InfiniBand.



Customizing Management of HPC Clusters

Thu, 01 February 2007 05:00:00 CST

High-performance computing (HPC) clusters are becoming increasingly widespread as many organizations' large-scale computing needs grow, but each type of cluster configuration comes with its own hardware management requirements. This article discusses best practices for customizing management software and hardware for HPC clusters based on DellTM PowerEdgeTM servers.



Using the Dell OpenManage Deployment Toolkit to Automate Configuration of HPC Cluster Nodes

Thu, 01 February 2007 05:00:00 CST

High-performance computing clusters comprising hundreds of nodes require one-to-many management utilities to configure the nodes with customized BIOS, baseboard management controller, and DellTM Remote Access Controller settings and to deploy Dell Utility Partitions in a reasonable amount of time. Administrators can use the Dell OpenManageTM Deployment Toolkit to perform these one-to-many operations across multiple nodes simultaneously.



Creating Scalable Compute Clusters Using the Sanbolic Melio File System and LaScala Volume Manager

Thu, 01 February 2007 05:00:00 CST

High-performance compute clusters can provide the performance and scalability necessary for processing-intensive applications. This article discusses using the Sanbolic® Melio® File System and Sanbolic LaScala® volume manager to create cost-effective, scalable clusters based on Dell™ servers and storage and the Microsoft® Windows® OS.



Serial Attached SCSI Storage for High-Performance Computing

Wed, 01 Nov 2006 05:00:00 CST

Serial Attached SCSI (SAS) is the successor of SCSI technology, and is designed to meet the performance requirements of new-generation servers by taking advantage of serial connection to help improve performance and scaling. This article discusses deployment of SAS storage in high-performance computing clusters.



Using OpenFabrics InfiniBand for HPC Clusters

Wed, 01 Nov 2006 05:00:00 CST

The OpenFabrics Alliance was formed to resolve issues with hardware and software interoperability and to deliver open source software for Remote Direct Memory Access fabric technologies like the InfiniBand architecture. Major initiatives by this alliance have resulted in the creation of an open source Linux® OS–based software stack for InfiniBand, which is rapidly gaining acceptance in the high-performance computing field.



Platform Open Cluster Stack: An Enhanced Cluster Software Package for Dell HPC Platforms

Wed, 01 Nov 2006 05:00:00 CST

Platform™ Open Cluster Stack (OCS) is an open source, standards-based cluster software stack designed to help administrators efficiently deploy and manage high-performance computing clusters. This article discusses some of the features and enhancements introduced in the latest Platform OCS version.



Secure HPC Cluster Management with Ninth-Generation Dell PowerEdge Servers

Wed, 01 Nov 2006 05:00:00 CST

The increasing use of high-performance computing (HPC) clusters for critical or sensitive high-performance computations has created a need for secure node management and monitoring capabilities. This article discusses the available tools for creating a secure HPC environment with the Dell OpenManage™ suite and ninth-generation Dell™ PowerEdge™ servers.



Evaluating Scalability and Power Benefits of Ninth-Generation Dell PowerEdge Servers in an HPC Environment

Wed, 01 Nov 2006 05:00:00 CST

Energy efficiency and scalability have become increasingly important to many enterprises. This article discusses the benefits of Intel® Xeon® 51xx processors for high-performance computing cluster environments by comparing performance/watt and cluster scalability results using compute-intensive applications.



Deploying Microsoft Windows Compute Cluster Server 2003 on Dell PowerEdge Servers

Wed, 01 Nov 2006 05:00:00 CST

Microsoft® Windows® Compute Cluster Server 2003 (CCS) can help provide a simple, cost-effective way to deploy and manage clusters. This article discusses CCS installation and configuration on Dell™ PowerEdge™ 1950 servers.



Configuring Ninth-Generation Dell PowerEdge Servers for High-Performance Computing Environments

Tue, 1 Aug 2006 05:00:00 CST

Dell engineers tested performance settings for ninth-generation Dell™ PowerEdge™ servers using representative compu-tation- and memory-intensive applications. Benchmark performance under various BIOS settings and memory configura-tions was also measured.



Remote Cluster Management Using Dell OpenManage Tools

Thu, 1 Feb 2006 05:00:00 CST

The Dell



Building One of the World’s Largest Windows-Based HPC Clusters

Thu, 26 June 2008 05:00:00 CST

When the University of Nebraska at Omaha’s Holland Computing Center teamed up with Dell to build a high-performance computing (HPC) cluster, the result was a flexible dual-boot system of Dell™ PowerEdge™ servers with dual-core AMD Opteron™ processors—and one of the world’s largest and most powerful Microsoft® Windows® OS–based clusters.



Accelerated HPC Productivity with Intel Cluster Ready Solutions

Tue, 7 Oct 2008 05:00:00 CST

The Intel® Cluster Ready program provides a standardized, replicable way to build and run high-performance computing (HPC) clusters, helping simplify cluster deployment and management. By using Intel Cluster Ready–certified Dell™ HPC clusters, organizations can quickly install and configure clusters to begin running registered HPC applications.



Building the Dell Supercomputer: New Dell Servers, Intel Cluster Ready, and Clustercorp Rocks+

Thu, 18 Jun 2009 05:00:00 CST

Designing and deploying a high-performance computing (HPC) cluster can be difficult for even the most experienced IT staff. To help simplify cluster acquisition and management, Dell, Intel, and Clustercorp have collaborated to offer an integrated and tested HPC cluster platform based on 11th-generation Dell™ PowerEdge™ servers, Intel® Cluster Ready configurations, and Clustercorp® Rocks+™ cluster management software.New 11th-generation Dell PowerEdge servers incorporating the quad-core Intel Xeon® processor 5500 series feature performance, energy-efficiency, and systems-management enhancements that make them well suited for HPC environments. Intel Cluster Ready–compliant configurations using these servers provide solutions that can run registered HPC applications right out of the box and do not require special cluster expertise to deploy and maintain. Rocks+, meanwhile, provides a robust, comprehensive software stack for HPC clusters, and can incorporate additional components such as the Dell Roll and Intel Cluster Ready Roll for simplified cluster deployment, the Mellanox OpenFabrics Enterprise Distribution (OFED) Roll or QLogic Roll for InfiniBand-based environments, and the Moab Roll or Platform LSF Roll for workload and resource management.After a cluster has been deployed with Rocks+, the Intel Cluster Checker tool confirms that the cluster is Intel Cluster Ready compliant. Cluster installation and verification can take place at the deployment site or at the factory through the Dell Enterprise Factory Integration (EFI) program, in which the clusters are racked, cabled, installed, and verified at Dell. The clusters resulting from the collaboration between Dell, Intel, and Clustercorp are designed to be simple, comprehensive, and reliable—helping to avoid the inherent complexity of managing disparate components, simplify HPC cluster deployment and management, and enable administrators and researchers to focus on running their applications rather than on cluster tuning and management.