Dell High Performance Computing (HPC) Solutions Enable Research from Leading Higher Education, Government and Industry Labs
- Higher education and research organizations like the Universities of Colorado at Boulder, Kentucky and the NASA Center for Climate Simulation select Dell for significant computational power and data analysis capabilities
New Dell blades, servers, storage platforms and workstations are designed to increase performance and efficiency in HPC environments
Scientists and engineers are looking for innovative ways to accelerate applications and shorten computational cycles. Dell is helping them by strengthening its portfolio of PowerEdge servers, storage expansion platforms and integrated solutions designed to meet the demands of the most compute intensive environments in education, government and industry, including the energy, financial services and life sciences sectors.
Innovation Meeting Customer Needs
To help customers deploy complex data subsystems, and workstations that put incredible computational power into a user’s hands, Dell continues to invest heavily in and support innovative high performance computing (HPC) technologies that include GPGPU (General-Purpose computation on Graphics Processing Units) acceleration solutions, and storage offerings.
These new Dell HPC solutions include:
The PowerEdge C6105, a 2U, 4-system solution where each system is a 2-socket AMD-based server that delivers a low TCO and high power efficiency for scaled out applications in a maximum-density chassis. The C6105 features hot-swap server nodes and hard disk drives along with redundant power supplies making it ideal for energy and budget constrained scale-out environments.
The PowerEdge C410x, an external PCI Express (PCIe) expansion chassis that connects 1-8 host nodes to 1-16 PCIe devices to get results fast, while saving on space, weight and cost. The system does not have CPUs or memory but is designed for efficiently powering and cooling GPUs and other PCIe devices. External PCIe chassis provide more flexibility and density while saving power and reducing costs compared to internal server GPUs.
On the measure of single-precision floating point performance, a PowerEdge C410x PCIe expansion chassis loaded with 16 NVIDIA Tesla ™ M2050 GPU modules delivers up to 16.5 Teraflops of computing throughput.
The PowerEdge M610x, bringing a new set of capabilities to Dell’s blade portfolio providing customers database acceleration and expanded compute capabilities by integrating two x16 PCIe slots in a dense 2S blade form factor. Customers can take advantage of NVIDIA’s new Tesla class GPUs or Fusion-IO’s ioDrive Duo drives.
Dell is also offering the new high performance Tesla™ C2050 GPU computing processor on the Dell Precision™ T7500 workstation, transforming it into a personal super computer for general purpose scientific and engineering computing. The Dell Precision T7500 along with the Tesla C2050 provides an ideal platform for oil & gas, financial, life sciences, academic and governmental organizations running compute intensive CUDA applications.
Partner technologies and innovations, such as Platform Computing Cluster Manager 2.0.1, Dell Edition, which allows users to quickly deploy, run and manage heterogeneous Linux cluster s with ease. In addition, Dell is working with Terascala to introduce standardized storage platforms, targeted for customers utilizing Lustre® file systems.
Availability and Integration
All Dell HPC solutions have been tested and validated so users can trust that the solution is not just a sum of parts but a tightly integrated, ready-to-run platform. The newly validated solution stack including, Platform Cluster Manager, HPC Storage Solution, C2050 with the Precision C7500 workstations, PowerEdge C410x and M610x servers are available immediately while the C6105 will begin shipping in the near future.
This portfolio will support academic institutions like the University of Colorado at Boulder and The University of Kentucky as well as public and private sector research organizations committed to increasing HPC efficiency in order to conduct ground breaking research. Researchers at the University of Colorado at Boulder are using Dell HPC solutions to study areas of earth‐system science, biotechnology and renewable energy while at the NASA Center for Climate Simulation, studies are aimed at simulation modeling and data analysis to explore, understand and predict climate change.
Availability and Integration
Quotes Steve Conway, Research Vice President, High Performance Computing, IDC & Steering Committee Member, HPC User Forum (www.hpcuserforum.com)
“Dell has become one of the leading players in the worldwide HPC market by delivering increasingly dense, scalable platforms and storage systems to buyers in the government, academic and industrial sectors. IDC expects the HPC market to exhibit healthy growth over the next five years, including new areas such as cloud computing and high-end business analytics, and Dell is well positioned to benefit from this growth.”
Steve Conway, Research Vice President, High Performance Computing, IDC & Steering Committee Member, HPC User Forum (www.hpcuserforum.com)
“When we debuted the NASA Center for Climate Simulation this spring, we anticipated expanding our high-performance computing capabilities to meet the growing needs of NASA’s earth and space science user community. We chose to use Dell PowerEdge servers based upon both the commitment of Dell to the HPC community and the ability of their systems to scale with our program over time.”
Phil Webster, Chief, NASA Goddard Space Flight Center's Computational and Information Sciences and Technology Office
About Dell: ###
Dell (NASDAQ: DELL) listens to its customers and uses that insight to make technology simpler and create innovative solutions that deliver reliable, long-term value. Learn more at www.dell.com.
 Based on information provided by NVIDIA at NVIDIA Tesla M2050 Fermi cards. Dell makes no warranties or representations regarding the accuracy of this information. Any questions or comments regarding this information should be addressed to NVIDIA. Actual performance will vary based on configuration, usage and manufacturing variability.