China Extends Lead in Number of TOP500 Supercomputers, US Holds on to Performance Advantage

BERKELEY, Calif.; FRANKFURT, Germany; and KNOXVILLE, Tenn.— The 54th edition of the TOP500 saw China and the US maintaining their dominance of the list, albeit in different categories. Meanwhile, the aggregate performance of the 500 systems, based on the High Performance Linpack (HPL) benchmark, continues to rise and now sits at 1.65 exaflops. The entry level to the list has risen to 1.14 petaflops, up from 1.02 petaflops in the previous list in June 2019.

Read more

News Feed

Looking Ahead To Marvell’s Future ThunderX Processors

If you are going to take on Intel in server processors, you have to play the same kind of long game that Intel itself played as it jumped from the desktop to the datacenter.

Looking Ahead To Marvell’s Future ThunderX Processors was written by Timothy Prickett Morgan at The Next Platform.

HDR 200Gb/s InfiniBand Sees Growth on November 2019 TOP500 Supercomputers List

BEAVERTON, Ore., Dec. 10, 2019 – The InfiniBand Trade Association (IBTA), a global organization dedicated to maintaining and furthering the InfiniBand specification, today reported the latest TOP500 List results show that HDR 200Gb/s InfiniBand accelerates 31 percent of new InfiniBand-based systems on the List, including the fastest TOP500 supercomputer built this year. The results also highlight InfiniBand’s continued […]

The post HDR 200Gb/s InfiniBand Sees Growth on November 2019 TOP500 Supercomputers List appeared first on HPCwire.

Supercomputers Help Predict Carbon Dioxide Levels

The Earth’s terrestrial ecosystems – its lands, forests, jungles and so on – are crucial “sinks” for atmospheric carbon, holding nearly 30 percent of our annual CO2 emissions as they breathe in the carbon-rich gas and convert it into biomass growth. However, widespread land-use change and deforestation has introduced severe uncertainty when estimating future carbon […]

The post Supercomputers Help Predict Carbon Dioxide Levels appeared first on HPCwire.

GigaIO Wins Most Innovative New Product Award for Big Data

Today GigaIO announced that the company's FabreX technology has been selected as the winner of Connect's Most Innovative New Product Award for Big Data. The Most Innovative New Product Awards is an annual competition that recognizes San Diego industry leaders for their groundbreaking contributions to technology and life sciences sectors. "FabreX is a cutting-edge network architecture that drives the performance of data centers and high-performance computing environments. Featuring a unified, software-driven composable infrastructure, the fabric dynamically assigns resources to facilitate streamlined application deployment, meeting today’s growing demands of data-intensive programs such as Artificial Intelligence and Deep Learning. FabreX adheres to industry standard PCI Express (PCIe) technology and integrates computing, storage and input/output (IO) communication into a single-system cluster fabric for flawless server-to-server communication. Optimized with GPU Direct RDMA (GDR) and NVMe-oF, FabreX facilitates direct memory access by a server to the system memories of all other servers in the cluster, enabling native host-to- host communication to create the industry’s first in-memory network."

The post GigaIO Wins Most Innovative New Product Award for Big Data appeared first on insideHPC.

Verne Global brings hpcDIRECT to Bonseyes AI Marketplace

Today Verne Global announced that is has partnered with the Bonseyes secure marketplace and collaborative platform for building and trading AI applications. The move will provide Bonseyes users with access to Verne Global’s hpcDIRECT optimized bare-metal infrastructure. "By collaborating with Verne Global, the Bonseyes marketplace users will be able to turn their competitive AI concepts into reality, knowing it is supported by a first-class, HPC-optimized data center and a skilled team of engineers and advisors."

The post Verne Global brings hpcDIRECT to Bonseyes AI Marketplace appeared first on insideHPC.

Fugaku Remakes Exascale Computing In Its Own Image

When originally conceived, Japan’s Post-K supercomputer was supposed to be the country’s first exascale system.

Fugaku Remakes Exascale Computing In Its Own Image was written by Michael Feldman at The Next Platform.

RIKEN's Satoshi Matsuoka provides appetizing teasers on upcoming Fugaku at HPC User Forum

At the 75th HPC User Forum in Edinburgh, Scotland, Satoshi Matsuoka, director of RIKEN talked about Fugaku, the next generation supercomputer in Japan. RIKEN is a big national laboratory, such as Max Planck in Germany. It is one of the 15 centres in Japan. There are other centres such as the AI centre, as well as centres for material sciences, biology, brain studies and regenerative medicine but RIKEN is the HPC Center and national flagship Research Center..

D-Wave expands Leap API into Amazon Braket

D-Wave Systems Inc., an expert in quantum computing systems, software, and services, has expanded the Leap API into Amazon Braket, a new, fully managed Amazon Web Services (AWS) solution that allows scientists, researchers, and developers to begin experimenting with computers from quantum hardware providers in a single place. The service, announced at AWS re:Invent 2019, gives AWS users live Cloud access to D-Wave's 2000Q quantum processor via the Leap quantum Cloud service API. Customers and developers will be able to leverage D-Wave's processing power in real time to run quantum applications directly from Amazon Braket. This access will extend to forthcoming D-Wave systems, such as the next-generation Advantage, coming in mid-2020..

The List

11/2019 Highlights

Since June 2019 only Petaflop systems have been able to make the list. The total aggregate performance of all 500 system has now risen to 1.65 Exaflops.

Two IBM build systems called Summit and Sierra and installed at DOE’s Oak Ridge National Laboratory (ORNL) in Tennessee and Lawrence Livermore National Laboratory in California kept the first two positions in the TOP500 in the USA.  

The share of installations in China continues to rise strongly. 45.6 % of all system are now listed as being installed in China. The share of system listed in the USA remains near it's all time low at 23.4 %. However, systems in the USA are on average larger, which allowed the USA (37.1%) to stay close to China (32.3%) in terms of installed performance.

There were no changes to the top of the list at all. The first new system shows up only at position 24! It is an IBM Power based system utilizing NVidia Volta GV100 which allowed it to capture the No 3 spot on the Green500 list.


List Statistics



Summit - IBM Power System AC922, IBM POWER9 22C 3.07GHz, NVIDIA Volta GV100, Dual-rail Mellanox EDR Infiniband
Sierra - IBM Power System S922LC, IBM POWER9 22C 3.1GHz, NVIDIA Volta GV100, Dual-rail Mellanox EDR Infiniband
IBM / NVIDIA / Mellanox
Sunway TaihuLight - Sunway MPP, Sunway SW26010 260C 1.45GHz, Sunway
Tianhe-2A - TH-IVB-FEP Cluster, Intel Xeon E5-2692v2 12C 2.2GHz, TH Express-2, Matrix-2000
Frontera - Dell C6420, Xeon Platinum 8280 28C 2.7GHz, Mellanox InfiniBand HDR
Dell EMC
Piz Daint - Cray XC50, Xeon E5-2690v3 12C 2.6GHz, Aries interconnect , NVIDIA Tesla P100
Cray Inc.
Trinity - Cray XC40, Xeon E5-2698v3 16C 2.3GHz, Intel Xeon Phi 7250 68C 1.4GHz, Aries interconnect
Cray Inc.
AI Bridging Cloud Infrastructure (ABCI) - PRIMERGY CX2570 M4, Xeon Gold 6148 20C 2.4GHz, NVIDIA Tesla V100 SXM2, Infiniband EDR
SuperMUC-NG - ThinkSystem SD650, Xeon Platinum 8174 24C 3.1GHz, Intel Omni-Path
Lassen - IBM Power System S922LC, IBM POWER9 22C 3.1GHz, Dual-rail Mellanox EDR Infiniband, NVIDIA Tesla V100
IBM / NVIDIA / Mellanox