New GPU-Accelerated Supercomputers Change the Balance of Power on the TOP500

The 51st (25th anniversary) Top500 list was released June 25, 2018 in Frankfurt Germany.
Here's some of Top500's thinking on the latest list and the state of the art.

From Top500, June 26:

For the first time in history, most of the flops added to the TOP500 list came from GPUs instead of CPUs. Is this the shape of things to come?

In the latest TOP500 rankings announced this week, 56 percent of the additional flops were a result of NVIDIA Tesla GPUs running in new supercomputers – that according to the Nvidians, who enjoy keeping track of such things. In this case, most of those additional flops came from three top systems new to the list: Summit, Sierra, and the AI Bridging Cloud Infrastructure (ABCI).

Summit, the new TOP500 champ, pushed the previous number one system, the 93-petaflop Sunway TaihuLight, into second place with a Linpack score of 122.3 petaflops. Summit is powered by IBM servers, each one equipped with two Power9 CPUs and six V100 GPUs. According to NVIDIA, 95 percent of the Summit’s peak performance (187.7 petaflops) is derived from the system’s 27,686 GPUs.

NVIDIA did a similar calculation for the less powerful, and somewhat less GPU-intense Sierra, which now ranks as the third fastest supercomputer in the world at 71.6 Linpack petaflops. And, although very similar to Summit, it has four V100 GPUs in each dual-socked Power9 node, rather than six. However, the 17,280 GPUs in Sierra still represent the lion’s share of that system’s flops.

Likewise for the new ABCI machine in Japan, which is now that country’s speediest supercomputer and is ranked fifth in the world. Each of its servers pairs two Intel Xeon Gold CPUs with four V100 GPUs. Its 4,352 V100s deliver the vast majority of the system’s 19.9 Linpack petaflops.

As dramatic as that 56 percent number is for new TOP500 flops, the reality is probably even more impressive.  According to Ian Buck, vice president of NVIDIA's Accelerated Computing business unit, more than half the Tesla GPUs they sell into the HPC/AI/data analytics space are bought by customers who never submit their systems for TOP500 consideration. Although many of these GPU-accelerated machines would qualify for a spot on the list, these particular customers either don’t care about all the TOP500 fanfare or would rather not advertise their hardware-buying habits to their competitors.

It’s also worth mentioning that the Tensor Cores in the V100 GPUs, with their specialized 16-bit matrix math capability, endow these three new systems with more deep learning potential than any previous supercomputer. Summit alone boasts over three peak exaflops of deep learning performance. Sierra’s performance in this regard is more in the neighborhood of two peak exaflops, while the ABCI number is around half an exaflop. Taken together, these three supercomputers represent more deep learning capability than the other 497 systems on the TOP500 list combined, at least from the perspective of theoretical performance.

The addition of AI/machine learning/deep learning into the HPC application space is a relatively new phenomenon, but the V100 appears to be acting as a catalyst. “This year’s TOP500 list represents a clear shift towards systems that support both HPC and AI computing,” noted TOP500 author Jack Dongarra, Professor at University of Tennessee and Oak Ridge National Lab....MORE

Working backward, here is the Top500 press release announcing the new list, June 25:

US Regains TOP500 Crown with Summit Supercomputer, Sierra Grabs Number Three Spot 

FRANKFURT, Germany; BERKELEY, Calif.; and KNOXVILLE, Tenn.—The TOP500 celebrates its 25^th anniversary with a major shakeup at the top of the list. For the first time since November 2012, the US claims the most powerful supercomputer in the world, leading a significant turnover in which four of the five top systems were either new or substantially upgraded.

Summit, an IBM-built supercomputer now running at the Department of Energy’s (DOE) Oak Ridge National Laboratory (ORNL), captured the number one spot with a performance of 122.3 petaflops on High Performance Linpack (HPL), the benchmark used to rank the TOP500 list. Summit has 4,356 nodes, each one equipped with two 22-core Power9 CPUs, and six NVIDIA Tesla V100 GPUs. The nodes are linked together with a Mellanox dual-rail EDR InfiniBand network.

Sunway TaihuLight, a system developed by China’s National Research Center of Parallel Computer Engineering & Technology (NRCPC) and installed at the National Supercomputing Center in Wuxi, drops to number two after leading the list for the past two years. Its HPL mark of 93 petaflops has remained unchanged since it came online in June 2016.

Sierra, a new system at the DOE’s Lawrence Livermore National Laboratory took the number three spot, delivering 71.6 petaflops on HPL. Built by IBM, Sierra’s architecture is quite similar to that of Summit, with each of its 4,320 nodes powered by two Power9 CPUs plus four NVIDIA Tesla V100 GPUs and using the same Mellanox EDR InfiniBand as the system interconnect.

Tianhe-2A, also known as Milky Way-2A, moved down two notches into the number four spot, despite receiving a major upgrade that replaced its five-year-old Xeon Phi accelerators with custom-built Matrix-2000 coprocessors. The new hardware increased the system’s HPL performance from 33.9 petaflops to 61.4 petaflops, while bumping up its power consumption by less than four percent. Tianhe-2A was developed by China’s National University of Defense Technology (NUDT) and is installed at the National Supercomputer Center in Guangzhou, China.

The new AI Bridging Cloud Infrastructure (ABCI) is the fifth-ranked system on the list, with an HPL mark of 19.9 petaflops. The Fujitsu-built supercomputer is powered by 20-core Xeon Gold processors along with NVIDIA Tesla V100 GPUs. It’s installed in Japan at the National Institute of Advanced Industrial Science and Technology (AIST).

Piz Daint (19.6 petaflops), Titan (17.6 petaflops), Sequoia (17.2 petaflops), Trinity (14.1 petaflops), and Cori (14.0 petaflops) move down to the number six through 10 spots, respectively.

General highlights
Despite the ascendance of the US at the top of the rankings, the country now claims only 124 systems on the list, a new low. Just six months ago, the US had 145 systems. Meanwhile, China improved its representation to 206 total systems, compared to 202 on the last list. However, thanks mainly to Summit and Sierra, the US did manage to take the lead back from China in the performance category. Systems installed in the US now contribute 38.2 percent of the aggregate installed performance, with China in second place with 29.1 percent. These numbers are a reversal compared to six months ago.
The next most prominent countries are Japan, with 36 systems, the United Kingdom, with 22 systems, Germany with 21 systems, and France, with 18 systems. These numbers are nearly the same as they were on the previous list....MORE

And the list:

June 2018
The TOP500 celebrates its 25^th anniversary with a major shakeup at the top of the list. For the first time since November 2012, the US claims the most powerful supercomputer in the world, leading a significant turnover in which four of the five top systems were either new or substantially upgraded....MUCH MORE

Earlier today:
"NVIDIA Chief Scientist Bill Dally on How GPUs Ignited AI, and Where His Team’s Headed Next" (NVDA)

High Performance Computing: "Europe’s First Exascale Supercomputer Powers Up" (plus the Top500 list of the world's fastest computers)

This is the big leagues in computing and they are tackling the big questions: complex-chaotic systems and our old nemesis, turbulence.

From IEEE Spectrum, June 26:

Debuting at No. 4 on the TOP500, JUPITER could open vast scientific vistas

If you want to visualize the surface air flowing across every meter of the planet Earth, you need to plot nearly three trillion squares of land and sea and space. To do that you need a mind-bogglingly big calculator.

Which is why Ioan Hadade, a computational scientist working with vast weather forecasting and climate models, is excited about the machine now online an hour down the road from his lab in Bonn, Germany. Europe’s first exascale supercomputer—called JUPITER, after a much bigger planet than our own—is nearly fully operational. It is currently running scientific programs on its formidable processors.

JUPITER debuted at No. 4 in the June 2025 global TOP500 list of the world’s most powerful computer systems.

It is based at the Jülich Supercomputing Center in the German Rhineland between Cologne and Aachen, running on a booster module with 5,900 accelerating compute nodes. Some 24,000-odd Nvidia Grace-Hopper superchips give JUPITER its oomph; the machine also features a universal cluster module with 1,300 nodes using Rhea1 processors, and an InfiniBand NDR network for the high-speed interconnects.

The semi-annual TOP500 rankings are a way to engage every single element of a machine for performance. Benchmarking proves the functionality of a highly complex operation. “And now, it’s better to have some science done on the machine,” says Thomas Lippert, director of the Jülich Supercomputing Center.

Computational Science at Scale
As of mid-June research enterprises were on the JUPITER machine testing scientific calculations. “You need a really large machine to run this,” Hadade says. He’s referring to the Destination Earth digital twin projects he and his colleagues are part of developing at the European Center for Medium-Range Weather Forecasts....

....MUCH MORE 

Here's the press release from Top500:

News 
El Capitan Retains Top Spot in 65th TOP500 List as Exascale Era Expands

The 65th edition of the TOP500 showed that the El Capitan system retains the No. 1 position. With El Capitan, Frontier, and Aurora, there are now 3 Exascale systems leading the TOP500. All three are installed at Department of Energy (DOE) laboratories in the United States.

The El Capitan system at the Lawrence Livermore National Laboratory, California, remains the No. 1 system on the TOP500. The HPE Cray EX255a system was measured with 1.742 EFlop/s on the HPL benchmark. LLNL now also submitted a measurement for the HPCG benchmark, achieving 17.41 Petaflop/s, which makes the system the new No. 1 on this ranking as well.

El Capitan has 11,039,616 cores and is based on AMD 4th generation EPYC processors with 24 cores at 1.8 GHz and AMD Instinct MI300A accelerators. It uses the HPE Slingshot interconnect for data transfer and achieves an energy efficiency of 60.3 Gigaflops/watt. El Capitan is the 3rd system exceeding the Exaflop mark on the HPL benchmark.

The Frontier system at the Oak Ridge National Laboratory, Tennessee, is the No. 2 system on the TOP500. Frontier has been remeasured with an HPL score of 1.353 EFlop/s.  

Frontier is based on the HPE Cray EX235a architecture and is equipped with AMD 3rd generation EPYC 64C 2GHz processors. The system has 8,699,904 total cores and also relies on HPE Slingshot interconnect for data transfer.  

The Aurora system at the Argonne Leadership Computing Facility, Illinois, was submitted with 1.012 EFlop/s on the HPL benchmark, which keeps it in the No. 3 spot on the TOP500. 

Aurora is built by Intel based on the HPE Cray EX - Intel Exascale Compute Blade, which uses Intel Xeon CPU Max Series processors and Intel Data Center GPU Max Series accelerators, which communicate through HPE Slingshot interconnect.

The JUPITER Booster system at the EuroHPC / Jülich Supercomputing Centre in Germany at No. 4 is the only new system in the TOP 10. 

JUPITER - JU Pioneer for Innovative and Transformative Exascale Research – was announced as the first EuroHPC exascale supercomputer (see https://jupiter.fz-juelich.de). It is currently being commissioned and has achieved a preliminary HPL value of 793.4 Petaflop/s on a partial system. The system is located at the Forschungszentrum Jülich campus in Germany and is operated by the Jülich Supercomputing Centre. It is based on the Eviden’s BullSequana XH3000 direct liquid-cooled architecture.

Here is a summary of the system in the Top 10....

....MUCH MORE 

The Fastest Computers In The World: Four Of The Ten Fastest Now Based In Europe

Plus two corporately-owned machines in the top ten.

First up, from Top500.com, May 13:

Highlights - June 2024

This is the 63rd edition of the TOP500.

The 63rd edition of the TOP500 reveals that Frontier has once again claimed the top spot, despite no longer being the only exascale machine on the list. Additionally, a new system has found its way into the Top 10.

The Frontier system at Oak Ridge National Laboratory in Tennessee, USA remains the most powerful system on the list with an HPL score of 1.206 EFlop/s. The system has a total of 8,699,904 combined CPU and GPU cores, an HPE Cray EX architecture that combines 3rd Gen AMD EPYC CPUs optimized for HPC and AI with AMD Instinct MI250X accelerators, and it relies on Cray’s Slingshot 11 network for data transfer. On top of that, this machine has an impressive power efficiency rating of 52.59 GFlops/Watt – putting Frontier at the No. 11 spot on the GREEN500.

Also like the last list, the Aurora system at the Argonne Leadership Computing Facility in Illinois, USA, has claimed the No. 2 spot on the TOP500. Despite currently being commissioned and not fully complete, Aurora is now the second machine to officially break the exascale barrier with an HPL score of 1.012 EFlop/s – an improvement over the 585.34 PFlop/s score from the last list. This system is based on HPE Cray EX- Intel Exascale Computer Blade and uses Intel Xeon CPU Max series processors, Intel Data Center GPU Max Series accelerators, and a Slingshot-11 interconnect.

The Eagle system installed on the Microsoft Azure Cloud in the USA reclaimed the No. 3 spot that it achieved after its debut appearance on the previous list, and it remains the highest-ranking cloud system on the TOP500. This Microsoft NDv5 system has an HPL score of 561.2 PFlop/s and is based on Intel Xeon Platinum 8480C processors and NVIDIA H100 accelerators.

Fugaku also retained its No. 4 spot from the previous list, despite holding the No.1 spot from June 2020 until November 2021. Based in Kobe, Japan, Fugaku has an HPL score of 442 PFlop/s and it remains the highest-ranked system outside the USA. The LUMI system at EuroHPC/CSC in Finland also remained in its spot at No. 5 with an HPL score of 380 PFlop/s. This machine is the largest system in Europe.

The only new system to find its way onto the Top 10 is the Alps machine at No. 6 from the Swiss National Supercomputing Centre (CSCS) in Switzerland. This system achieved an HPL score of 270 PFlop/s.

Here is a summary of the systems in the Top 10:....

....MUCH MORE

The corporate machines in the top ten are:

#3 Microsoft Azure United States  

#10 NVIDIA Corporation United States 

Nvidia dominates in accelerator chips while Intel has a majority of the CPU business:

Highlights from the List

A total of 194 systems on the list are using accelerator/co-processor technology, up from 185 six months ago. 83 of these use NVIDIA Ampere chips, 48 use NVIDIA Volta, and 22 systems with 18.

		Count	System Share (%)	Rmax (TFlops)	Rpeak (TFlops)	Cores
1	NVIDIA Tesla V100	34	6.8	160,352	258,591	3,042,576
2	NVIDIA A100	28	5.6	334,074	501,370	3,381,552
3	NVIDIA A100 SXM4 40 GB	17	3.4	182,940	256,548	1,863,032
4	NVIDIA A100 SXM4 80 GB	12	2.4	69,404	81,030	622,400
5	NVIDIA Tesla A100 80G	11	2.2	125,864	167,923	1,088,032
6	AMD Instinct MI250X	10	2	1,753,739	2,480,336	12,660,800
7	NVIDIA Tesla V100 SXM2	10	2	88,521	177,436	1,997,584
8	NVIDIA Tesla A100 40G	8	1.6	58,849	94,137	616,484
9	NVIDIA GH200 Superchip	7	1.4	472,151	663,724	2,401,488
10	NVIDIA H100	7	1.4	637,532	969,686	2,238,328

....MUCH MORE 

And the rest of the Top500 package:

TOP500 Release

• The List
• Press Release
• List highlights

Statistics

• Performance Development
• Sublist Generator
• List Statistics
• Tree Maps
• Historical Charts

Downloads

• TOP500 List (XML)
• TOP500 List (Excel)
• GREEN500 List (Excel)
• TOP500 Poster
• Poster in PDF

And as we said a few days ago, the Green500 list was released at the same time as the Top500. 

And more to come.

China Has Fallen Out of the Top 10 Fastest Supercomputer Rankings; Europe Now Has Four

From Top500.org, November 2024:

News
El Capitan achieves top spot, Frontier and Aurora follow behind

The 64th edition of the TOP500 reveals that El Capitan has achieved the top spot and is officially the third system to reach exascale computing after Frontier and Aurora. Both systems have since moved down to No. 2 and No. 3 spots, respectively. Additionally, new systems have found their way onto the Top 10.

The new El Capitan system at the Lawrence Livermore National Laboratory in California, U.S.A., has debuted as the most powerful system on the list with an HPL score of 1.742 EFlop/s. It has 11,039,616 combined CPU and GPU cores and is based on AMD 4th generation EPYC processors with 24 cores at 1.8GHz and AMD Instinct MI300A accelerators. El Capitan relies on a Cray Slingshot 11 network for data transfer and achieves an energy efficiency of 58.89 Gigaflops/watt. This power efficiency rating helped El Capitan achieve No. 18 on the GREEN500 list as well.

The Frontier system at Oak Ridge National Laboratory in Tennessee, U.S.A, has moved down to the No. 2 spot. It has increased its HPL score from 1.206 Eflop/s on the last list to 1.353 Eflop/s on this list. Frontier has also increased its total core count, from 8,699,904 cores on the last list to 9,066,176 cores on this list. It relies on Cray’s Slingshot 11 network for data transfer.

The Aurora system at Argonne Leadership Computing Facility in Illinois, U.S.A, has claimed the No. 3 spot on this TOP500 list. The machine kept its HPL benchmark score from the last list, achieving 1.012 Exaflop/s. Aurora is built by Intel based on the HPE Cray EX – Intel Exascale Compute blade which uses Intel Xeon CPU Max Series Processors and Intel Data Center GPU Max Series accelerators that communicate through Cray’s Slingshot-11 network interconnect.

The Eagle system installed on the Microsoft Azure Cloud in the U.S.A. claimed the No. 4 spot and remains the highest-ranked cloud-based system on the TOP500. It has an HPL score of 561.2 PFlop/s

The only other new system in the TOP 5 is the HPC6 system at No. 5. This machine is installed at Eni S.p.A center in Ferrera Erbognone, Italy and has the same architecture as the No. 2 system Frontier. The HPC6 system at Eni achieved an HPL benchmark of 477.90 PFlop/s and is now the fastest system in Europe.

Here is a summary of the system in the Top 10:

• The El Capitan system at the Lawrence Livermore National Laboratory, California, USA is the new No. 1 system on the TOP500. The HPE Cray EX255a system was measured with 1.742 Exaflop/s on the HPL benchmark. El Capitan has 11,039,616 cores and is based on AMD 4th generation EPYC™ processors with 24 cores at 1.8 GHz and AMD Instinct™ MI300A accelerators. It uses the Cray Slingshot 11 network for data transfer and achieves an energy efficiency of 58.89 Gigaflops/watt.

• Frontier is now the No. 2 system in the TOP500. This HPE Cray EX system was the first US system with a performance exceeding one Exaflop/s. It is installed at the Oak Ridge National Laboratory (ORNL) in Tennessee, USA, where it is operated for the Department of Energy (DOE). It currently has achieved 1.353 Exaflop/s using 8,699,904 cores. The HPE Cray EX architecture combines 3rd Gen AMD EPYC™ CPUs optimized for HPC and AI, with AMD Instinct™ 250X accelerators, and a Slingshot-11 interconnect.

• Aurora is currently the No. 3 with a preliminary HPL score of 1.012 Exaflop/s. It is installed at the Argonne Leadership Computing Facility, Illinois, USA, where it is also operated for the Department of Energy (DOE). This new Intel system is based on HPE Cray EX - Intel Exascale Compute Blades. It uses Intel Xeon CPU Max Series processors, Intel Data Center GPU Max Series accelerators, and a Slingshot-11 interconnect.

• Eagle the No. 4 system is installed by Microsoft in its Azure cloud. This Microsoft NDv5 system is based on Xeon Platinum 8480C processors and NVIDIA H100 accelerators and achieved an HPL score of 561 Petaflop/s.

• The new No. 5 system is called HPC6 and installed at Eni S.p.A center in Ferrera Erbognone in Italy. It is another HPE Cray EX235a system with 3rd Gen AMD EPYC™ CPUs optimized for HPC and AI, with AMD Instinct™ 250X accelerators, and a Slingshot-11 interconnect. It achieved 477.9 Petaflop/s.

 ....MUCH MORE

And The List:

TOP500 List - November 2024

Milestone: China Builds The (NEW) World's Fastest Supercomuter Using Only Chinese Components (and other news) INTC; NVDA; IBM

China has had the world's fastest computer for the last three years or so, the Tianhe-2, which used Intel microprocessors so this latest computer is a remarkable achievement. The U.S. plans to recapture the top spot for the first time in five years years when Oak Ridge builds their latest machine using IBM CPUs, NVIDIA GPUs and NVIDIA's NV Link tying it all together. The ORNL 'puter should hit either the Nov. 2017 or June 2018 Top 500 lists.

Here's more of the story in three parts. First up, the Daily Signal:

China Builds World’s Fastest Computer

On Monday, the inconceivable happened. China announced it had built the world’s fastest computer. China has always been good at copying and/or stealing intellectual property, but it has rarely produced “indigenous innovation,” particular in the high-tech sector.

The Chinese supercomputer, called the Tianhe-1A, is capable of performing over 2.5 thousand trillion operations a second and is big enough to fill a large warehouse. The processors weigh over 150 tons and can store information equal to about a hundred million books.

At its peak, the computer can perform around 93,000 trillion calculations per second. Purportedly, the Chinese supercomputer is 30 percent faster than the fastest American computer. “Considering that just 10 years ago, China claimed a mere 28 systems on TOP500 global supercomputer listing, with none ranked in the top 30, the nation has come further and faster than any other country in the history of supercomputing,” said Top500.

Last year the U.S. blocked Intel from shipping faster semiconductor chips to China on national security grounds. According to the New York Times, the United States blocked the sale of advanced microprocessors to China over concerns they were being used in nuclear weapon development. Without the Intel chips, the Chinese were forced to develop their own semiconductors, which apparently they are doing.

Pierre Ferragu, an industry technical analyst, said the new rankings showed that China was “pulling together all the building blocks of an independent semiconductor value chain.”...

And the lead story at Top 500 (along with the new semi-annual rankings), June 20, 2016:

New Chinese Supercomputer Named World’s Fastest System on Latest TOP500 List
System achieves 93 petaflop/s running LINPACK on Chinese-designed CPUs China draws Equal to the U.S . in Overall Installations

FRANKFURT, Germany; BERKELEY, Calif.; and KNOXVILLE, Tenn.—China maintained its No. 1 ranking on the 47th edition of the TOP500 list of the world’s top supercomputers, but with a new system built entirely using processors designed and made in China. Sunway TaihuLight is the new No. 1 system with 93 petaflop/s (quadrillions of calculations per second) on the LINPACK benchmark.

Developed by the National Research Center of Parallel Computer Engineering & Technology (NRCPC) and installed at the National Supercomputing Center in Wuxi, Sunway TaihuLight displaces Tianhe-2, an Intel-based Chinese supercomputer that has claimed the No. 1 spot on the past six TOP500 lists.

The newest edition of the list was announced Monday, June 20, at the 2016 International Supercomputer Conference in Frankfurt. The closely watched list is issued twice a year.

Sunway TaihuLight, with 10,649,600 computing cores comprising 40,960 nodes, is twice as fast and three times as efficient as Tianhe-2, which posted a LINPACK performance of 33.86 petaflop/s. The peak power consumption under load (running the HPL benchmark) is at 15.37 MW, or 6 Gflops/Watt.

This allows the TaihuLight system to grab one of the top spots on the Green500 in terms of the Performance/Power metric.  Titan, a Cray XK7 system installed at the Department of Energy’s (DOE) Oak Ridge National Laboratory, is now the No. 3 system. It achieved 17.59 petaflop/s.

Rounding out the Top 10 are Sequoia, an IBM BlueGene/Q system installed at DOE’s Lawrence Livermore National Laboratory; Fujitsu’s K computer installed at the RIKEN Advanced Institute for Computational Science (AICS) in Kobe, Japan; Mira, a BlueGene/Q system installed at DOE’s Argonne National Laboratory; Trinity, a Cray X40 system installed at DOE/NNSA/LANL/SNL; Piz Daint, a Cray XC30 system installed at the Swiss National Supercomputing Centre  and the most powerful system in Europe; Hazel Hen, a Cray XC40 system installed at HLRS in Stuttgart, Germany; and Shaheen II, a Cray XC40 system installed at King Abdullah University of Science and Technology (KAUST) in Saudi Arabia is at No. 10.

The latest list marks the first time since the inception of the TOP500 that the U.S is not home to the largest number of systems. With a surge in industrial and research installations registered over the last few years, China leads with 167 systems and the U.S. is second with 165. China also leads the performance category, thanks to the No. 1 and No. 2 systems....MORE

More detail at Top 500's "China Tops Supercomputer Rankings with New 93-Petaflop Machine".

And from Forbes, June 17, a story we decided to hold until the Top500 list came out:

What To Expect At Next Week's ISC Supercomputing Event

Next week I will attend the annual international supercomputing event (now renamed the ISC High Performance Conference) in Frankfurt, Germany. This conference is the “tock” to the annual US-based “tick” Supercomputing event, which takes place every year after Thanksgiving and before Christmas. The European show is typically much smaller than its US cousin but affords attendees a close-up look into the vendors’ plans and the amazing science being conducted at global supercomputing centers and institutions. And it is always a good party, with some 3,000 attendees expected to make the trek to Frankfurt this year.

This will be the first ISC event to my knowledge where the keynote address is not about traditional High Performance supercomputing topics like simulation and modeling. This year, the keynote speaker is Andrew Ng, Chief Scientist at Baidu and associate professor at Stanford University. Andrew is a leading researcher in Artificial Intelligence (AI) and a high-profile advocate for the science of Machine Learning and Deep Neural Networks (DNN). This is a noteworthy departure from the norm, as the traditional High Performance Computing (HPC) community has a lot to gain from employing the techniques being researched and deployed by the DNN community and internet giants like Google, Amazon.com Facebook, Microsoft and others.

In addition to some awesome brews and brats, here are some topics I hope to learn more about at the show.

1. I expect we will see a status update on the upcoming Intel INTC “Knights Landing” multi-core Xeon Phi, which is expected to ship later this year. In addition to speeds and feeds, I’d like to see how it will compare to NVIDIA GPUs, especially the new Pascal generation of boards that will begin shipping about the same time. I am especially keen to learn about any benchmarks the company can provide for Deep Learning and to hear about Intel’s plans to invest in the Deep Learning Ecosystem.
2. It is also about time that we hear more from Intel on their plans for Altera FPGAs, especially as it relates to HPC and Deep Learning. Will it target training for Deep Learning, and if so, how will the company position FPGAs with respect to Xeon Phi?
3. From NVIDIA, I will want to hear about the productization of the Pascal P100 chip in Tesla products and also about the company’s plans for the inference side of Deep Learning outside of the automotive and embedded space where they have already mastered the market with the DrivePX2 platform. Specifically, I’d like to hear how the company plans to compete with the Google Tensor Processor for cloud AI services....

...MORE 

The Fiftieth TOP500 List of the Fastest Supercomputers in the World

We'll be back with more commentary on what's up in High Performance Computing but as a placeholder here's Top500 with the highest of the HPC crowd:

November 2017 

The fiftieth TOP500 list of the fastest supercomputers in the world has China overtaking the US in the total number of ranked systems by a margin of 202 to 143. It is the largest number of supercomputers China has ever claimed on the TOP500 ranking, with the US presence shrinking to its lowest level since the list’s inception 25 years ago.

Just six months ago, the US led with 169 systems, with China coming in at 160. Despite the reversal of fortunes, the 143 systems claimed by the US gives them a solid second place finish, with Japan in third place with 35, followed by Germany with 20, France with 18, and the UK with 15.

China has also overtaken the US in aggregate performance as well. The Asian superpower now claims 35.4 percent of the TOP500 flops, with the US in second place with 29.6 percent.

The top 10 systems remain largely unchanged since the June 2017 list, with a couple of notable exceptions.

Sunway TaihuLight, a system developed by China’s National Research Center of Parallel Computer Engineering & Technology (NRCPC), and installed at the National Supercomputing Center in Wuxi, maintains its number one ranking for the fourth time, with a High Performance Linpack (HPL) mark of 93.01 petaflops.

Tianhe-2 (Milky Way-2), a system developed by China’s National University of Defense Technology (NUDT) and deployed at the National Supercomputer Center in Guangzho, China, is still the number two system at 33.86 petaflops.

Piz Daint, a Cray XC50 system installed at the Swiss National Supercomputing Centre (CSCS) in Lugano, Switzerland, maintains its number three position with 19.59 petaflops, reaffirming its status as the most powerful supercomputer in Europe. Piz Daint was upgraded last year with NVIDIA Tesla P100 GPUs, which more than doubled its HPL performance of 9.77 petaflops.

The new number four system is the upgraded Gyoukou supercomputer, a ZettaScaler-2.2 system deployed at Japan’s Agency for Marine-Earth Science and Technology, which was the home of the Earth Simulator. Gyoukou was able to achieve an HPL result of 19.14 petaflops. using PEZY-SC2 accelerators, along with conventional Intel Xeon processors. The system’s 19,860,000 cores represent the highest level of concurrency ever recorded on the TOP500 rankings of supercomputers.

Titan, a five-year-old Cray XK7 system installed at the Department of Energy’s (DOE) Oak Ridge National Laboratory, and still the largest system in the US, slips down to number five. Its 17.59 petaflops are mainly the result of its NVIDIA K20x GPU accelerators.

Sequoia, an IBM BlueGene/Q system installed at DOE’s Lawrence Livermore National Laboratory, is the number six system on the list with a mark of 17.17 petaflops. It was deployed in 2011.

The new number seven system is Trinity, a Cray XC40 supercomputer operated by Los Alamos National Laboratory and Sandia National Laboratories. It was recently upgraded with Intel “Knights Landing” Xeon Phi processors, which propelled it from 8.10 petaflops six months ago to its current high-water mark of 14.14 petaflops.

Cori, a Cray XC40 supercomputer, installed at the National Energy Research Scientific Computing Center (NERSC), is now the eighth fastest supercomputer in the world. Its 1,630 Intel Xeon "Haswell" processor nodes and 9,300 Intel Xeon Phi 7250 nodes yielded an HPL result of 14.01 petaflops.

At 13.55 petaflops, Oakforest-PACS, a Fujitsu PRIMERGY CX1640 M1 installed at Joint Center for Advanced High Performance Computing in Japan, is the number nine system. It too is powered by Intel “Knights Landing” Xeon Phi processors.

Fujitsu’s K computer installed at the RIKEN Advanced Institute for Computational Science (AICS) in Kobe, Japan, is now the number 10 system at 10.51 petaflops. Its performance is derived from its 88 thousand SPARC64 processor cores linked by Fujitsu’s Tofu interconnect. Despite its tenth-place showing on HPL, the K Computer is the top-ranked system on the High-Performance Conjugate Gradients (HPCG) benchmark....MUCH MORE, including links to all 500 of the fastest machines on earth.

There Is A New World's Fastest Supercomputer (maybe) NVDA

From the measurement geeks at Top500, May 30:

Highlights - June 2022

This is the 59th edition of the TOP500.

The 59th edition of the TOP500 revealed the Frontier system to be the first true exascale machine with an HPL score of 1.102 Exaflop/s.

We have a new No 1, the Frontier system at the Oak Ridge National Laboratory (ORNL), Tennessee, USA. Frontier brings the pole position back to the USA after it was held for 2 years by the Fugaku system at RIKEN Center for Computational Science (R-CCS) in Kobe, Japan. The Frontier system is currently being integrated and tested at ORNL. It has a peak performance of 1.6 ExaFlop/s and has achieved so far, an HPL benchmark score of 1.102 Eflop/s. On the HPL-AI benchmark, which measure performance for mixed precision calculation, Frontier already demonstrated 6.86 Exaflops!

We also have a new No. 3, the LUMI system at EuroHPC/CSC in Finland, and the largest system in Europe. The third newcomer to the top 10 is at No. 10, the Adastra system at GENCI-CINES in France.

All 3 new systems in the top 10 are based on the latest HPE Cray EX235a architecture, which combines 3rd Gen AMD EPYC™ CPUs optimized for HPC and AI with AMD Instinct™ 250X accelerators, and Slingshot interconnects.

Here is a summary of the system at the Top10:

• Frontier is the new No. 1 system in the TOP500. This HPE Cray EX system is the first US system with a peak performance exceeding one ExaFlop/s. It is currently being integrated and tested at the ORNL in Tennessee, USA, where it will be operated by the Department of Energy (DOE). It currently has achieved 1.102 Exaflop/s using 8,730,112 cores. The new HPE Cray EX architecture combines 3rd Gen AMD EPYC™ CPUs optimized for HPC and AI with AMD Instinct™ 250X accelerators and Slingshot-11 interconnect.
• Fugaku, now the No. 2 system, is installed at the RIKEN Center for Computational Science (R-CCS) in Kobe, Japan. It has 7,630,848 cores which allowed it to achieve an HPL benchmark score of 442 Pflop/s. This puts it 3x ahead of the No. 3 system in the list. 
• The new LUMI system, another HPE Cray EX system installed at EuroHPC center at CSC in Finland, is the new No. 3 with a performance of 151.9 Pflop/s just ahead of No 4. The European High-Performance Computing Joint Undertaking (EuroHPC JU) is pooling European resources to develop top-of-the-range Exascale supercomputers for processing big data. One of the pan-European pre-Exascale supercomputers, LUMI, is in CSC's data center in Kajaani, Finland.
• Summit, an IBM-built system at ORNL in Tennessee, USA, is now listed at the No. 4 spot worldwide with a performance of 148.8 Pflop/s on the HPL benchmark which is used to rank the TOP500 list. Summit has 4,356 nodes, each housing two Power9 CPUs with 22 cores and six NVIDIA Tesla V100 GPUs, each with 80 streaming multiprocessors (SM). The nodes are linked together with a Mellanox dual-rail EDR InfiniBand network.
• Sierra, a system at the Lawrence Livermore National Laboratory, CA, USA, is at No. 5. Its architecture is very similar to the #4 systems Summit. It is built with 4,320 nodes with two Power9 CPUs and four NVIDIA Tesla V100 GPUs. Sierra achieved 94.6 Pflop/s.

....MUCH MORE

Two European (LUMI and France's Adastra), two Chinese and Japan's Fugaku  add diversity to the U.S. national laboratories machines with NVIDIA's Selene being the only corporate machine in the top10. Germany and Italy have machines in the 11th and 12th fastest positions with Microsoft's Azure in the 13th spot. Here are the first 100 fastest.

The highlights press release (above) takes pains to point out how deeply NVDA accelerators have penetrated supercomputer architecture: 

Highlights from the List

A total of 170 systems on the list are using accelerator/co-processor technology, up from 151 six months ago. 84 of these use NVIDIA Volta chips, 54 use NVIDIA Ampere, and 8 systems with NVIDIA Pascal....

And here is the "maybe", from DataCenterDynamics, May 30:

Oak Ridge's exascale 'Frontier' system named world's most powerful supercomputer on Top500
But may still be behind a secret Chinese supercomputer

With the Summit Supercomputer, U.S. Could Retake Computing’s Top Spot (NVDA)

We've been babbling about this 'puter for three years, usually in the context of its use of NVIDIA GPU's and NV Link connections, more after the jump.

From IEEE Spectrum:

Oak Ridge’s 200-petaflop Summit supercomputer will come on line in mid-2018

In November of 2012, the semiannual Top500 rankings of the world’s supercomputers gave top billing to a machine constructed at the Oak Ridge National Laboratory, in Tennessee. Aptly named Titan, the machine boasted a peak performance of more than 27 × 10¹⁵ floating-point operations per second, or 27 petaflops. It was an immense computing resource for researchers in government, industry, and academe, and being at the top of the supercomputing heap, it helped to boost pride within the U.S. high-⁠performance computing community.

The satisfaction was short-lived. Just seven months later, Titan lost the world-supercomputing crown to a Chinese machine called Tianhe-2 (Milky Way-2). And three years on, yet another Chinese number-crunching behemoth—the Sunway TaihuLight—took over the title of world’s most powerful supercomputer. Its peak performance was 125 petaflops. After that, Titan wasn’t looking so titanic anymore.

Using the Sunway TaihuLight, Chinese researchers captured the 2016 Gordon Bell Prize [PDF] for their work modeling atmospheric dynamics. “That shows it wasn’t just a stunt machine,” says Jack Dongarra of the University of Tennessee, one of the creators of the Top500 rankings.

You might be wondering why for the past five years the United States has seemingly given up on reclaiming the top spot. In fact, there was no such surrender. In 2014, U.S. engineers drafted proposals for a new generation of supercomputers. The first of these will bear fruit later this year in the form of a supercomputer named Summit, which will replace Titan at Oak Ridge. The new machine’s peak performance will be around 200 petaflops when it comes on line in a few months, which will make it the most powerful supercomputer on the planet.
Maybe.

“We’re very open in the U.S. with our machines,” says Arthur “Buddy” Bland, project director of the Leadership Computing Facility at Oak Ridge. That is, he’s confident that Summit will be completed as planned and that it will be the most powerful supercomputer in the United States. But in the meantime, China, or some other country for that matter, could field a new supercomputer or upgrade an existing one to exceed Summit’s performance. Could that really happen? “We have no idea,” says Bland.

He and his colleagues at Oak Ridge aren’t losing any sleep over the question—and they need all the sleep they can get these days because they still have a lot of work ahead of them as they labor to replace Titan with Summit. They are not, however, following the pattern that they used to build Titan, which was created as a result of a series of increasingly elaborate upgrades to an earlier Oak Ridge supercomputer called Jaguar....MUCH MORE

Previously:
May 2016
NVIDIA Sets New All Time High On Pretty Good Numbers, "Sweeping Artificial Intelligence Adoption" (NVDA)

We are fans.
Before we go any further, our NVIDIA boilerplate: we make very few calls on individual names on the blog but this one is special. 

They are positioned to be the brains in autonomous vehicles, they will drive virtual reality should it ever catch on, the current businesses include gaming graphics, deep learning/artificial intelligence, and supercharging the world's fastest supercomputers including what will be the world's fastest at Oak Ridge next year. 

Not just another pretty face.
Or food delivery app....

Sept. 2017
"The Astonishing Engineering Behind America's Latest, Greatest Supercomputer"

...a) Wired does not mention the graphics processing units are from NVIDIA

b) the Chinese may have taken an insurmountable lead in the need-for-speed derby and the ORNL computer, designed to be the world's fastest may not make it.

The folks at Wired are smart and have been on the tech beat for a long time, they should know better than to do puff pieces.  

If interested in this stuff, whether for modeling complex-chaotic systems such as markets or weather or for national security applications or just because supercomputers are amazing in their own right see also:

April 2016

CERN Will Be Using NVIDIA Graphics Processors to Accelerate Their Supercomputer (NVDA)

June 2016

Milestone: China Builds The (NEW) World's Fastest Supercomuter Using Only Chinese Components (and other news) INTC; NVDA; IBM

China has had the world's fastest computer for the last three years or so, the Tianhe-2, which used Intel microprocessors so this latest computer is a remarkable achievement. The U.S. plans to recapture the top spot for the first time in five years years when Oak Ridge builds their latest machine using IBM CPUs, NVIDIA GPUs and NVIDIA's NV Link tying it all together. The ORNL 'puter should hit either the Nov. 2017 or June 2018 Top 500 lists....

November 15, 2016 

NVIDIA: Don't Buy the Stock For The Autonomous Car Stuff (or virtual reality) NVDA; TSLA; IBM

November 16, 2016

NVIDIA Builds Its Very Own Supercomputer, Enters The Top500 List At #28 (NVDA)

Now they're just showing off.

The computer isn't going to be a product line or anything that generates immediate revenues but it puts the company in a very exclusive club and may lead to some in-house breakthroughs in chip design going forward.
The stock is up $4.97 (+5.77%) at $91.16.

To be clear, this isn't someone using NVDA's graphics processors to speed up their supercomputer as the Swiss did with the one they let CERN use and which is currently the eighth fastest in the world or the computer that's being built right now at Oak Ridge National Laboratory and is planned to be the fastest in the world (but may not make it, China's Sunway TaihuLight is very, very fast).

And this isn't the DIY supercomputer we highlighted back in May 2015:

...Among the fastest processors in the business are the one's originally developed for video games and known as Graphics Processing Units or GPU's. Since Nvidia released their Tesla hardware in 2008 hobbyists (and others) have used GPU's to build personal supercomputers.

Here's Nvidias Build your Own page.
Or have your tech guy build one for you....

Nor is it the $130.000 supercomputer NVIDIA came up with for companies to get started in Deep Learning/AI.

No, this is NVIDIA's very own supercomputer.

Here's the brand new list (they come out every six months):
Top500 List - November 2016

July 22, 2017

Supercomputers "The 49th TOP500 List was published June 20, 2017 in Frankfurt, Germany." 

We're usually more timely posting the list but reality keeps intruding on the blog stuff.
A couple things to point out, we've made a few mentions of the Swiss supercomputer Piz Daint, here's one of them from last November:
NVIDIA Builds Its Very Own Supercomputer, Enters The Top500 List At #28 (NVDA)

...To be clear, this isn't someone using NVDA's graphics processors to speed up their supercomputer as the Swiss did with the one they let CERN use and which is currently the eighth fastest in the world or the computer that's being built right now at Oak Ridge National Laboratory and is planned to be the fastest in the world (but may not make it, China's Sunway TaihuLight is very, very fast)....

You can see the results of the upgrade in the current list, Piz Daint went from 8th fastest to 3rd fastest in the world. 

Possibly also of interest, NVIDIA's 'puter has been bumped down to #32, behind Facebook's AI/machine-learning supercomputer which is based on NVIDIA's DGX-1 and uses NVDA chips as their graphics accelerator....

In fact wary reader may have come to the conclusion that NVIDIA and supercomputers have become an idée fixe for yours truly. From 2 1/2 years ago:

May 2015
Nvidia Wants to Be the Brains Of Your Autonomous Car (NVDA)

We've mentioned, usually in the context of the Top 500* fastest supercomputers, that:

Long time readers know we have a serious interest in screaming fast computers and try to get to the Top500 list a couple times a year. Here is a computer that was at the top of that list, the fastest computer in the world just four years ago. And it's being shut down.
Technology changes pretty fast. 

That was from a 2013 post.

Among the fastest processors in the business are the one's originally developed for video games and known as Graphics Processing Units or GPU's. Since Nvidia released their Tesla hardware in 2008 hobbyists (and others) have used GPU's to build personal supercomputers.
Here's Nvidias Build your Own page.
Or have your tech guy build one for you.

In addition Nvidia has very fast connectors they call NVLink.
Using a hybrid combination of IBM Central Processing Units (CPU's) and Nvidia's GPU's, all hooked together with the NVLink, Oak Ridge National Laboratory is building what will be the world's fastest supercomputer when it debuts in 2018.

As your kid plays Grand Theft Auto....

*Here's the Top 500 site, the next list is due next month. China’s National University of Defense Technology has had the top spot since the June 2013 list when it toppled Oak Ridge National Laboratory's Titan.

Be all that as it may be, here's ORNL's webpage for the new supercomputer, Summit. 

Supercomputers: Oak Ridge National Lab Still Has The World's Fastest, Followed By Japan and Finland (but what about China?)

Okay, okay, "High Performance Computing" supercomputers.

From IEEE Spectrum, November 15:

Top500: Frontier Still No. 1. Where’s China?
Largely unchanged supercomputer rankings foreground U.S. efforts to ratchet down Chinese HPC ambitions

The latest list of the world’s most powerful supercomputers reveals that Frontier, at Oak Ridge National Lab, in Tennessee, has stayed on top. The newly released Top500 list could arguably be seen as a temporary object lesson in stasis, while still pointing toward future aspirants and aspiring countries who could one day challenge Frontier’s crown.

With a performance of 1.1 exaflops, or 1.1 quintillion floating-point operations per second, Frontier was the first machine to break the exascale barrier, a threshold of a billion billion calculations per second. It is still the only exascale supercomputer announced to date, according to this week’s ranking of the world’s fastest supercomputers.

Frontier, which is based on the latest HPE Cray EX235a architecture and boasts more than 8.7 million AMD cores, remains powerful enough to perform more than twice as well as the No. 2 machine, Fugaku, at the Riken Center for Computational Science, in Japan. Fugaku had led the Top500 list for two years until Frontier ousted it in June. Running more than 7.6 million Fujitsu cores, Fugaku’s performance, at 442 petaflops, joins Frontier in posting an unchanged compute speed since June’s Top500 list.

“Frontier is a first-of-a-kind supercomputer comprised of a hybrid architecture to run calculations at an unprecedented speed,” says Justin Whitt, program director for the U.S. Department of Energy’s Oak Ridge Leadership Computing Facility at Oak Ridge National Laboratory. “Our experienced team of technical staff and vendor partners worked tirelessly for Frontier to achieve the world’s first exascale performance on the Linpack benchmark as reported in May 2022.” Whitt citedthreeprojects that have used Frontier since its unveiling to earn finalist status for the Gordon Bell prize, to be awarded at the Supercomputing 2022 conference in Dallas this week.

At third place on the Top500 list is the Lumi system in Finland, which uses an HPE Cray EX235a architecture and harnesses some 2.2 million AMD cores. LUMI has doubled in power since June—with a performance of 309 petaflops—and remains the most powerful supercomputer in Europe....

....MUCH MORE

Top500 Supercomputers List, November 2025: "El Capitan retains Top500 crown as Jupiter Booster becomes Europe’s first exascale system"

From DatacenterDynamics, November 19:

AMD and Eviden selected for Alice Recoque exascale supercomputer 

El Capitan has retained its title as the world’s most powerful supercomputer on the most recent edition of the Top500 list, with the US Department of Energy once again operating all of the top three systems.

However, while all the supercomputers comprising the top ten retained their positions from six months ago, the fourth-place Jupiter Booster system officially became Europe’s first exascale supercomputer, posting an HPL score of exactly one exaflops – a significant increase from the 793.4 petaflops it achieved in June.

Housed at Germany's Forschungszentrum Jülich campus, the Jupiter Booster – a partial version of the larger supercomputer being built at the supercomputing facility – is a BullSequana XH3000 featuring Nvidia GH200 Superchips.

El Capitan also saw an increase in its performance for the 66th edition of the list, posting a four percent gain from six months earlier for a total of 1.809 exaflops. Housed at Lawrence Livermore, the HPE Cray EX255a-based system is comprised of AMD 4th Gen Epyc 24C 1.8GHz CPUs and Instinct MI300A GPUs.

The system is one of six HPE Cray systems to make the top ten, a list that also includes Frontier, Aurora, HPC6, Alps, and LUMI.

Achieving 1.353 exaflops, Oak Ridge's second-place Frontier system comprises AMD 3rd Generation Epyc 64C 2GHz CPUs and Instinct MI250X GPUs, while Argonne's Cray EX-based Aurora with the Intel Exascale Compute Blade, Xeon CPU Max 9470 52C 2.4GHz, and Intel Data Center GPU Max, had an HPL score of 1.012 exaflops.

In fifth place was Microsoft Azure's 561.2 petaflops Eagle, with Intel Xeon Platinum 8480C 48C 2GHz CPUs and Nvidia H100 GPUs, followed by the aptly named HPC6, another AMD-powered system operated by Italian oil giant Eni. The second most powerful supercomputer in Europe, HPC6, achieved 477.9 petaflops.

Japan’s Arm-based Fugaku supercomputer ranked seventh. Built by Fujitsu for Riken, it achieved 442.01 petaflops.

Eighth, ninth, and tenth place went to Switzerland's 434.9 petaflops Alps, Finland's 379.7 petaflops Lumi, and Italy's 241.2 petaflops Leonardo, respectively. Both Lumi and Leonardo form part of the European High-Performance Computing Joint Undertaking (EuroHPC JU) initiative.

Alice Recoque to become third AMD-powered exascale system
Following the release of the Top500 list, AMD and Eviden announced they would be building the Alice Recoque supercomputer, set to be Europe’s second exascale system....

....MORE

DCD home 

https://top500.org/ 

NVIDIA Builds Its Very Own Supercomputer, Enters The Top500 List At #28 (NVDA)

Now they're just showing off.

The computer isn't going to be a product line or anything that generates immediate revenues but it puts the company in a very exclusive club and may lead to some in-house breakthroughs in chip design going forward.
The stock is up $4.97 (+5.77%) at $91.16.

To be clear, this isn't someone using NVDA's graphics processors to speed up their supercomputer as the Swiss did with the one they let CERN use and which is currently the eighth fastest in the world or the computer that's being built right now at Oak Ridge National Laboratory and is planned to be the fastest in the world (but may not make it, China's Sunway TaihuLight is very, very fast).

And this isn't the DIY supercomputer we highlighted back in May 2015:

...Among the fastest processors in the business are the one's originally developed for video games and known as Graphics Processing Units or GPU's. Since Nvidia released their Tesla hardware in 2008 hobbyists (and others) have used GPU's to build personal supercomputers.

Here's Nvidias Build your Own page.
Or have your tech guy build one for you....

Nor is it the $130.000 supercomputer NVIDIA came up with for companies to get started in Deep Learning/AI.

No, this is NVIDIA's very own supercomputer.

Here's the brand new list (they come out every six months):
Top500 List - November 2016

And here is NVIDIA's 'puter, right behind one of the U.S. Army's machines and just ahead of Italian energy giant ENI's machine:

28

NVIDIA Corporation United States

DGX SATURNV - NVIDIA DGX-1, Xeon E5-2698v4 20C 2.2GHz, Infiniband EDR, NVIDIA Tesla P100 Nvidia

Some of our prior posts on the Top500.

Here's NVDA's story, from Extreme Tech:

Nvidia builds its own supercomputer, claims top efficiency spot in the TOP500

Every six months, the TOP500 team releases a list of the 500 most powerful supercomputers in the world based on their results in the long-running Linpack benchmark. These machines are typically owned by governments or are built as public-private partnerships between various government and industry partners who share costs and computer time. This year, Nvidia has made its own entry — and no, I don’t mean that Nvidia is powering someone else’s system, or that the company collaborated with a different firm. Nvidia built its own frickin’ supercomputer.

The new DGX SaturnV contains 60,512 CPU cores (the machine relies on Intel’s Xeon E5-2698v4 for its CPUs) and 63,488GB of RAM. The machine is actually a cluster of 125 DGX-1 systems — that’s the AI processing “supercomputer in a box” that Nvidia unveiled last year, and the first machine to feature the company’s Pascal GPUs (the full GP100 configuration). According to Nvidia, the new machine is 2.3x more energy efficient than the closest Xeon Phi system of equivalent performance and it delivers 9.46GFLOPS/watt, a 42% improvement over the most efficient system unveiled last June. That’s a huge improvement in a relatively short period of time, though I do want to note an important caveat to these kinds of figures. One thing we’ve covered before in our previous discussions of exascale computing is how ramping compute clusters upwards creates very different constraints than we typically consider when talking about desktops or even servers. Factors like interconnect power consumption, total memory loadout, and memory architecture all play a significant part in how these metrics play out.

In other words: Nvidia’s new performance/watt metrics are great for Pascal and a huge achievement, but I don’t think we can read much about the potential power efficiency of Xeon Phi without seeing something more closely akin to an apples-to-apples comparison. It’s also interesting that Nvidia chose to use Intel Xeons for its own power efficiency push than OpenPOWER, despite being a fairly vocal supporter of OpenPOWER and NVLink. Given that the new supercomputer relies on Nvidia’s DGX-1, however, it probably made more sense to build its own server clusters towards the x86 ecosystem rather than trying to launch a new AI and compute platform around Power at this time....MORE

Supercomputers: Why Aurora Didn't Take The #1 Position In The Latest Top500 List

It is a truly remarkable machine but the prime contractor, Intel, wasn't up to the job, delivered it very late and the machine is still being put together.

High performance computing from HPCWire, May 15:

Some Reasons Why Aurora Didn’t Take First Place in the Top500 List

The makers of the Aurora supercomputer, which is housed at the Argonne National Laboratory, gave some reasons why the system didn’t make the top spot on the Top500 list of the fastest supercomputers in the world.

At slightly over 1-exaflops of performance, Aurora remained in the second spot behind Frontier, which took the top spot at around 1.2 exaflops. Aurora passed the exaflop barrier, making it only the second system to pass the threshold.

The organizers seemed to prioritize the utility of the system over performance. Only time will tell whether the hundreds of millions of tax-payer money spent on the system was worth it.

ANL, HPE, and Intel, the main organizations behind the system, explained why Aurora isn’t complete and answered benchmarking questions.

Argonna runs on a chip Intel calls the “Exascale Compute Blade,” which has the 52-core Xeon CPU Max and Intel Data Center GPU Max.

The Benchmark Isn’t Complete
The Aurora system is still being installed, and more performance can be squeezed out of the system.

The high-performance LINPACK run represents about 80% to 90% of the overall system performance benchmarked and may still be able to topple Frontier to the top spot.

Aurora’s theoretical performance was estimated to be 2 exaflops. The real-world performance measure is about 55% of this, which is in line with the numbers achieved on other supercomputers, which are about 50% to 70%.

The System Wasn’t Built for High-performance LINPACK Runs
The hardware choices for Aurora indicate the system wasn’t built expressly to achieve top high-performance LINPACK benchmarks.

Instead, the system was built to balance scientific and AI computing. The system achieved 10.6 exaflops on mixed-precision computing on limited system benchmarking.

Scientific computing is shifting toward mixed-precision computing, and ANL, HPE, and Intel are looking ahead with Aurora. At the same time, Aurora meets the needs of conventional scientific applications that require double-precision computing.

The Aurora builders deliberately decided not to include processing units that drive up the main Top500 benchmark.

The System Has More AI Hardware Parts
Aurora’s design decision was to dedicate more silicon space and power budget to AI and mixed-precision parts than Frontier.

For example, Intel’s GPU, called Ponte Vecchio, in Aurora does not have dedicated matrix engines for FP64 (double-precision computing). By comparison, AMD’s MI250X in Frontier has dedicated parts for faster and more power-efficient FP64 matrix math calculations....

....MUCH MORE

Recently, on the lists:

• Jim Simons' Simons Foundation Is Running The World's Greenest Supercomputer
• Okay, One Last NVDA Post (this week): "Watt a Win: NVIDIA Sweeps New Ranking of World’s Most Energy-Efficient Supercomputers"
• The Fastest Computers In The World: Four Of The Ten Fastest Now Based In Europe

And previously on the Aurora machine:

• "The Computer That Will Change Everything"
• The Computer That Will Change Everything: "New Breed of Supercomputer Aims for the Two Quintillion Mark"

Supercomputers: Oak Ridge National Laboratory May Be First To 1,000,000,000,000,000,000 operations per second

From IEEE Spectrum, Dec. 23: 

Frontier Supercomputer to Usher in Exascale Computing 

In 2018, a new supercomputer called Summit was installed at Oak Ridge National Laboratory, in Tennessee. Its theoretical peak capacity was nearly 200 petaflops—that’s 200 thousand trillion floating-point operations per second. At the time, it was the most powerful supercomputer in the world, beating out the previous record holder, China’s Sunway TaihuLight, by a comfortable margin, according to the well-known Top500 ranking of supercomputers. (Summit is currently No. 2, a Japanese supercomputer called Fugaku having since overtaken it.)

In just four short years, though, demand for supercomputing services at Oak Ridge has outstripped even this colossal machine. “Summit is four to five times oversubscribed,” says Justin Whitt, who directs ORNL’s Leadership Computing Facility. “That limits the number of research projects that can use it.”

The obvious remedy is to get a faster supercomputer. And that’s exactly what Oak Ridge is doing. The new supercomputer being assembled there is called Frontier. When complete, it will have a peak theoretical capacity in excess of 1.5 exaflops.

The remarkable thing about Frontier is not that it will be more than seven times as powerful as Summit, stunning as that figure is. The remarkable thing is that it will use only twice the power. That’s still a lot of power—Frontier is expected to draw 29 megawatts, enough to power a town the size of Cupertino, Calif. But it’s a manageable amount, both in terms of what the grid there can supply and what the electricity bill will be.... 

....MUCH MORE

Previously on Summit:
January 11, 2018
With the Summit Supercomputer, U.S. Could Retake Computing’s Top Spot (NVDA)
We've been babbling about this 'puter for three years, usually in the context of its use of NVIDIA GPU's and NV Link connections, more after the jump....
June 8, 2018
"IBM builds world’s most powerful supercomputer to crack AI" (IBM; NVDA)
First rate, focusing not just on the raw speed but the AI capabilities as well....
June 28, 2018
New GPU-Accelerated Supercomputers Change the Balance of Power on the TOP500
The 51st (25th anniversary) Top500 list was released June 25, 2018 in Frankfurt Germany.
Here's some of Top500's thinking on the latest list and the state of the art....

And many more

"US launches Genesis Mission, training AI with government datasets to further science"

From DatacenterDynamics, November 25:

Will use DOE supercomputers and the cloud 

The US White House has launched the Genesis Mission, a plan to feed scientific federal government datasets into artificial intelligence models.

The Department of Energy (DOE)-led project is meant to create and train AI agents and scientific foundation models to speed up scientific progress.

"We will harness for the benefit of our nation the revolution underway in computing, and build on decades of innovation in semiconductors and high-performance computing," President Donald Trump said in an executive order launching the initiative.

"The Genesis Mission will dramatically accelerate scientific discovery, strengthen national security, secure energy dominance, enhance workforce productivity, and multiply the return on taxpayer investment into research and development, thereby furthering America’s technological dominance and global strategic leadership."

The program will use DOE supercomputers, cloud-based AI computing environments, and other Federal computing resources.

The order calls for the exact compute, storage, and networking resources to be made available within 90 days, including "resources available through industry partners."

Within 120 days, the DOE Secretary is expected to identify a set of initial data and model assets, including digitization, standardization, metadata, and provenance tracking. They will also develop a plan to ingest datasets from federally funded research, other agencies, academic institutions, and approved private-sector partners.

Then within 240 days, the Secretary will review capabilities across the DOE national laboratories and other participating Federal research facilities for robotic laboratories and production facilities "with the ability to engage in AI-directed experimentation and manufacturing, including automated and AI-augmented workflows and the related technical and operational standards needed."....

....MORE

The U.S. does have some resources. 

Way back in June 2018 we were posting "New GPU-Accelerated Supercomputers Change the Balance of Power on the TOP500". 

Here's the top 10 of the latest Top500 list, the world's three fastest supercomputers are at Department of Energy laboratories as well as numbers 12 and 15.

Top500 Supercomputers List, November 2025: "El Capitan retains Top500 crown as Jupiter Booster becomes Europe’s first exascale system"

The New Top500 List Of The World's Fastest Supercomputers Is Out (AMD; NVDA)

From Top500, June 28:

Fugaku Holds Top Spot, Exascale Remains Elusive

FRANKFURT, Germany; BERKELEY, Calif.; and KNOXVILLE, Tenn.— The 57 ^th edition of the TOP500 saw little change in the Top10. The only new entry in the Top10 is the Perlmutter system at NERSC at the DOE Lawrence Berkeley National Laboratory. The machine is based on the HPE Cray "Shasta" platform and a heterogeneous system with both GPU-accelerated and CPU-only nodes. Perlmutter achieved 64.6 Pflop/s, putting the supercomputer at No. 5 in the new list.

The Japanese supercomputer Fugaku held onto the top spot on the list. A system codeveloped by Riken and Fujitsu, Fugaku has an HPL benchmark score of 442 Pflop/s. This performance exceeds the No. 2 Summit by 3x. The machine is based on Fujitsu's custom ARM A64FX processor. What's more, in single or further reduced precision, which is often used in machine learning and AI, Fugaku's peak performance is actually above an exaflop. Such an achievement has caused some to introduce this machine as the first "Exascale" supercomputer. Fugaku already demonstrated this new level of performance on the new HPL-AI benchmark with 2 Eflop/s.

Outside of this, we saw quite a few instances of Microsoft Azure and Amazon EC2 Cloud instances fairly high on the list. Pioneer-EUS, the machine to snag the No. 24 spot and the No.27 Pioneer-WUS2, rely on Azure. The Amazon EC2 Instance Cluster at No. 41 utilizes Amazon EC2.

Here is a summary of the systems in the Top10:

• Fugaku remains the No. 1 system. It has 7,630,848 cores which allowed it to achieve an HPL benchmark score of 442 Pflop/s. This puts it 3x ahead of the No. 2 system in the list.
• Summit, an IBM-built system at the Oak Ridge National Laboratory (ORNL) in Tennessee, USA, remains the fastest system in the U.S. and at the No. 2 spot worldwide with a performance of 148.8 Pflop/s on the HPL benchmark, which is used to rank the TOP500 list. Summit has 4,356 nodes, each housing two Power9 CPUs with 22 cores each and six NVIDIA Tesla V100 GPUs, each with 80 streaming multiprocessors (SM). The nodes are linked together with a Mellanox dual-rail EDR InfiniBand network.
• Sierra, a system at the Lawrence Livermore National Laboratory, CA, USA is at No. 3. Its architecture is very similar to the #2 system Summit. It is built with 4,320 nodes with two Power9 CPUs and four NVIDIA Tesla V100 GPUs. Sierra achieved 94.6 Pflop/s.
• Sunway TaihuLight, a system developed by China's National Research Center of Parallel Computer Engineering & Technology (NRCPC) and installed at the National Supercomputing Center in Wuxi, which is in China's Jiangsu province, is listed at the No. 4 position with 93 Pflop/s.
• Perlmutter at No. 5 is new in the TOP10. It is based on the HPE Cray "Shasta" platform, and a heterogeneous system with AMD EPYC based nodes and 1536 NVIDIA A100 accelerated nodes. Perlmutter achieved 64.6 Pflop/s.
• Selene, now at No. 6, is an NVIDIA DGX A100 SuperPOD installed inhouse at NVIDIA in the USA. The system is based on an AMD EPYC processor with NVIDIA A100 for acceleration and a Mellanox HDR InfiniBand as a network and achieved 63.4 Pflop/s....

....MUCH MORE