Leading UK HPC Research Center Selects 232 Nodes for Upgrade to New Code-Based Rockport Switchless Network Architecture Showing Up to 28% Performance Improvement Over InfiniBand
By conquering congestion – a workload killer – Durham is able to accelerate search results with more predictability, better resource utilization and economic efficiency.
Rockport Joins ExCALIBUR Technology Supplier Panel
OTTAWA, Ontario, May 24, 2022–(BUSINESS WIRE)–Rockport Networks today announced that Institute for Computational Cosmology (ICC) at Durham University has selected the Rockport Switchless Network as part of the ExCALIBUR program around new network technologies for its COSMA7 supercomputer at DiRAC HPC facility. For the first time, the COSMA7 cluster and its 232 Rockport nodes will provide Durham University, DiRAC and the ExCALIBUR program with insight into the benefits of reduced congestion when modeling exascale workloads and using code to run on future exascale systems.
With Rockport’s switchless architecture, Durham maximizes its investment in HPC infrastructure by overcoming switch-intensive networking trade-offs while enabling elastic scalability and simplifying operations required for demanding workloads. in performance. Rockport’s fully distributed network fabric provides better performance, better resource utilization, and network economics by overcoming the bottlenecks associated with traditional switch-based networks.
ICC’s massively parallel research into dark matter and energy, black holes, planet formation and collisions requires enormous computing power where interconnection traditionally becomes a limiting factor. Interconnect-related performance and congestion issues encountered when handling advanced computing workloads lead to unpredictable execution times and underutilization of expensive compute and storage resources.
After an initial proof-of-concept deployment of Rockport in the Durham Intelligent NIC Environment (DINE) supercomputer, the university quickly saw Rockport’s technology as a revolutionary way to solve congestion, delivering faster load working under a load equal to or greater than that of InfiniBand.
“We are always on the lookout for advanced technologies that can improve the performance and reliability of the advanced computing workloads we run,” said Dr. Alastair Basden, DiRAC/Durham University, Technical Director of COSMA HPC Cluster. “Based on the results and our first experience with Rockport’s switchless architecture, we were confident in our choice to improve our exascale modeling performance – all supported by the right economics.”
The COSMA7 Rockport installation is supported by new funding from DiRAC and the Excalibur Hardware & Enabling Software (H&ES) program which assesses next generation high performance technologies for the highest priority areas of research in the UK.
COSMA7 helps scientists analyze the greatest mysteries of space, including dark energy, black holes and the origins of the universe. By deploying the Rockport Switchless Network in COSMA7, ICC researchers and scientists around the world can experience the power of Rockport’s architecture, avoid research delays, and better equip them to create complex simulations of the universe. Rockport provides deep insights into code performance and faster time-to-results on traditional switch-based networks.
The Rockport Switchless network distributes the network switching function to its COSMA7 endpoints (nodes) which become the network. By eliminating layers of switches, the Rockport network ensures that compute and storage resources no longer run out of data, and researchers have more predictability regarding workload runtime.
Redesigning network switches creates an opportunity to take advantage of direct interconnect topologies that provide a connectivity mesh where each network endpoint can efficiently forward traffic to all other endpoints. The Rockport Switchless Network is a distributed, highly reliable, high performance interconnect providing pre-wired supercomputer topologies through a standard plug-and-play Ethernet interface.
Durham particularly liked the simplicity and intuitive look inside the network provided by Rockport’s Autonomous Network Manager (ANM). ANM provides a holistic view and deep insight into the active network. ICC sees end-to-end traffic flows and can troubleshoot network performance in near real-time – a level of transparency that has long been lacking in the industry.
“The expansion of our network within ICC’s DiRAC HPC cluster is very exciting for our team as it has proven that there is no need to compromise on performance and throughput,” said said Matthew Williams, CTO, Rockport Networks. “Tackling congestion goes beyond provisioning additional switches to solve the bandwidth problem. Sophisticated control and architecture means the customer is no longer at the mercy of bottlenecks created by their network infrastructure. »
Rockport Joins ExCALIBUR Technology Supplier Panel
Rockport has also joined the Technology Providers Panel (TPP) formed to enable ExCALIBUR’s H&ES program to strategically interact and collaborate with technology providers to increase the effectiveness and efficiency of their projects.
About Rockport Networks
The next generation of high-performance networking from Rockport Networks frees up the entire data center to deliver more results, faster, and with greater economic and environmental sustainability. Inspired by the world’s fastest supercomputers, the Rockport Switchless Network replaces centralized switch architectures with a high-performance distributed direct interconnect that self-discovers, configures and repairs, and is simple and transparent to use. By virtually eliminating congestion and latency, data center workloads can run much faster, enabling organizations to improve ROI and make critical decisions faster. Learn more about www.rockportnetworks.com and follow us on Twitter and LinkedIn.
DiRAC stands for Distributed Research Using Advanced Computing. The DiRAC high-performance computing facility provides state-of-the-art supercomputing resources to UK researchers working on cutting-edge scientific computing across a wide range of fields, including particle physics, astrophysics, cosmology and nuclear physics. It includes supercomputers in Cambridge, Durham, Leicester and Edinburgh, each designed to support specific types of calculations. DiRAC also provides access to a team of expert research software engineers to help researchers make the best use of available computing resources. For more details, see: https://dirac.ac.uk/.
About the ICC
The Institute for Computational Cosmology is a world center for research in cosmology and astrophysics using large supercomputers. It hosts around 50 researchers who, together with collaborators around the world, are responsible for some of the largest and most sophisticated cosmological simulations in the world. These are used to answer fundamental questions regarding the nature of dark and dark energy, as well as the formation of galaxies and other large cosmological structures. To see https://www.icc.dur.ac.uk
About ExCALIBUR H&ES
ExCALIBUR (Exascale Computing Algorithms and Infrastructures Benefitting UK Research) is a UK £45.7 million Strategic Priorities Fund (SPF) program led by the Met Office, UKAEA and UKRI which aims to provide the next generation of high performance simulation software for the highest priority areas of UK research. ExCALIBUR enables the redesign of high priority computer codes and algorithms to meet the demands of advanced technology and UK research. Under the ExCALIBUR initiative, UKRI is investing £4.5 million over five years to create and evaluate new pre-Exascale Hardware and Enabling Software (H&ES) testbeds, in a collaboration of co-design with technology partners and the UK research community.
See the source version on businesswire.com: https://www.businesswire.com/news/home/20220524005070/en/
Rockport Networks Inc.
Phone number: +1 (613) 608-6417
Separate communications for Rockport Networks Inc.
Mobile: +1 (917) 575-5830