FRANKFURT, Germany; BERKELEY, Calif.; and KNOXVILLE, Tenn.—The 55th edition of the TOP500 saw some significant additions to the list, spearheaded by a new number one system from Japan. The latest rankings also reflect a steady growth in aggregate performance and power efficiency.
Find the highlights of the 55th TOP500 list below, presented by Erich Strohmaier. Slides from the session are also available now.
Detailed report on the Fujitsu Fugaku system.
Certificates for the top ranking systems in the TOP500 list, Green500 list, HPCG and HPL-AI benchmarks.
Data centers host many users and applications and have become the competitive advantage for research organization and manufacturing companies. Keeping the data center intact and healthy is critical, as operational costs of supercomputers continue to rise, driven by growing scientific computing demands and new security threats. Moreover, malicious users may exploit data center access to misuse compute resources by running prohibited applications, such as crypto currency mining, resulting in unexpected downtimes and higher operating costs. This week, NVIDIA unveiled the Unified Fabric Manager (UFM) Cyber-AI platform, which minimizes downtime and saves OPEX in InfiniBand data centers by harnessing AI-powered analytics …
Sometimes, vendors just won’t take customers’ money for a cluster no matter how hard the customer pleads. Last month, HPC technology strategist Ryan Quick, co-founder and principal at Providentia Worldwide consulting firm, told us that while vendors have technologies capable of delivering solutions across the stack, too often they have disconnects between business units that makes buying an integrated solution difficult, if not impossible. In this interview, Quick expands on those comments with problem scenarios and advice on handling vendors.
The post Prying an Integrated HPC Cluster from Vendors’ Hands: ‘We Had to Force Them to Get into the Same Room’ appeared first on insideHPC.
San Diego Supercomputer Center makes high performance computing resources available to researchers via a “condo cluster” model. Many homebuyers have found that the most affordable path to homeownership leads to a condominium, in which the purchaser buys a piece of a much larger building. This same model is in play today in the high performance […]
The post High performance computing — at condo prices appeared first on HPCwire.
August 10, 2020 — “It is often said that a decade in computer science is a century in other sciences,” said Xun (Steve) Jian, an assistant professor in the Department of Computer Science, of the aspects of his field that fascinate him. “New computer hardware technologies constantly emerge.” “Such a dynamic field provides many new problems to […]
The post Virginia Tech Professor Receives NSF Career Award to Bolster Cyberinfrastructure Performance appeared first on HPCwire.
UK colocation provider Kao Data talks about the changing data center requirements of HPC organizations in the “Innovation Corridor,” between London and Cambridge. Kao Data Vice President Spencer Lamb talks with the late Rich Brueckner about his company’s adoption of hyperscalers’ infrastructure principles to offer state-of-the-art, high density data centers. For more insights from Kao […]
The post Kao Data Talks HPC-class Colocation Data Center Requirements in the UK’s ‘Innovation Corridor’ appeared first on insideHPC.
Multiplying things by two and putting them on a roadmap is easy, even if it does take a lot of courage to do that. …
The Tech Tricks That Make PCI-Express 6.0 And Beyond Possible was written by Timothy Prickett Morgan at The Next Platform.
On today’s program we look at the future of Cray as part of HPE with former Cray CEO and current HPC and mission-critical systems lead at HPE-Cray, Pete Ungaro. …
Next Platform TV for August 6, 2020 was written by Nicole Hemsoth at The Next Platform.
The new top system, Fugaku, turned in a High Performance Linpack (HPL) result of 415.5 petaflops, besting the now second-place Summit system by a factor of 2.8x. Fugaku, is powered by Fujitsu’s 48-core A64FX SoC, becoming the first number one system on the list to be powered by ARM processors. In single or further reduced precision, which are often used in machine learning and AI applications, Fugaku’s peak performance is over 1,000 petaflops (1 exaflops). The new system is installed at RIKEN Center for Computational Science (R-CCS) in Kobe, Japan.
The most energy-efficient system on the Green500 is the MN-3, based on a new server from Preferred Networks. It achieved a record 21.1 gigaflops/watt during its 1.62 petaflops performance run. The system derives its superior power efficiency from the MN-Core chip, an accelerator optimized for matrix arithmetic. It is ranked number 395 in the TOP500 list.
In second position is the new NVIDIA Selene supercomputer, a DGX A100 SuperPOD powered by the new A100 GPUs. It occupies position seven on the TOP500.