#Dell #EMC now has a Top500 #Supercomputer in its #HPC Innovation Lab for customers and partners to use. The #Zenith system is designed to complement Dell EMC efforts to advance state-of-the-art HPC solutions from the workgroup to the TOP500. The lab’s goal is to provide a place where organizations can test new technologies, collaborate with seasoned engineers, and see the benefits that can be realized when making use of a system that integrates the latest innovations in compute, storage, and networking. Zenith also lets Dell EMC and #Intel demonstrate the performance and power that can be realized for the most demanding workloads. Driving the need for change Commercial enterprises are changing their HPC requirements. Compute demands, sophistication of analytics and simulations routines, and the volume of data needing analysis or visualization keep growing. While the workloads are different, the HPC problems are similar in wide-ranging industries such as energy exploration, manufacturing, the life sciences, and finance. Workflows can be impacted by any number of factors. As a result, organizations of all sizes find the need for HPC resources that rival that found in government labs and academic supercomputer centers. In particular, to meet today’s demands organization have a plethora of technology choices and numerous hardware and software elements that must be selected, tightly integrated, optimized, and managed over time. Unfortunately, most organizations do not have the time, budget, or internal expertise in exotic technologies to pull this off. Also unfortunate, there are no standard configuration options or best practices that ensure a system will meet the specific needs of a business. Because of the great intricacies of selecting the best technologies and bringing the various elements of an HPC system together, organizations can greatly benefit by trying their workloads on a real system. This will allow companies to get an understanding of which technologies and hardware and software combinations deliver the performance and reliability for their specific business-critical workloads. That is exactly what Dell EMC has in mind with the work being done at its HPC Innovation Lab. Whenever Dell EMC investigates a new HPC technology, they bring it into the lab to comprehend its impact on the system and performance. The lab explores the design, development, and integration of HPC systems, with a focus on the software stack, plus compute, interconnect, and storage performance analysis and performance. That work is then shared with customers and partners. To evaluate technologies for the most demanding commercial workloads, the lab hosts Dell EMC’s Zenith HPC system. Ranked at 372 on the recently released Top500 list of the world’s most powerful supercomputers, Zenith is designed on Intel’s Scalable Systems Framework (SSF). It is a 451-teraflops (Linpack) machine built with Dell PowerEdge C6320 and PowerEdge R630 servers using Xeon E5-2697v4 processors and the Intel Omni-Path interconnect. Dell EMC will also be unveiling a companion system (not yet submitted to the TOP500), Rattler, that has 80 C6320 PowerEdge nodes fully connected with EDR. Dell EMC uses Zenith to prototype and characterize performance of advanced technologies for general HPC use and specifically for several vertical markets. Zenith is used for co-development with partners and customer evaluation of software for scalability and performance. Zenith has also been used to create proof of concepts blending HPC and cloud technologies, HPC and #Big Data analytics, OpenStack distributions for HPC, Hadoop analytics framework distribution running on a Lustre database, and many more. Leveraging years of HPC expertise The work at Dell EMC’s HPC Innovation Lab builds on years of hands-on expertise in delivering systems for some of the most challenging commercial HPC workloads, as well as supplying some of the world’s most powerful supercomputers to academic research centers and government labs. Dell EMC HPC and supercomputing systems include Jetstream at Indiana University Pervasive Technology Institute, Comet at the San Diego Supercomputer Center, and the Stampede systems at Texas Advanced Computing Center. The Zenith HPC system brings the technology lessons learned in these and other systems to commercial workloads. As a result, Dell EMC is one of the only, if not the only Tier1 vendor to build a supercomputer for customer/partner use. For more information about the Zenith HPC system, the Dell EMC HPC Innovation Lab, and Dell EMC’s HPC efforts visit: http://www.dellhpc.org/ or http://www.dell.com/hpc.
https://www.hpcwire.com/2016/12/05/taking-supercomputing-enterprise/
No comments:
Post a Comment