Analysis As a sign of how great AMD’s resurgence in high-performance computing has become, the latest list of the world’s 500 fastest publicly known supercomputers shows that the chip designer has become a favorite among organizations implementing x86-based HPC clusters.
The most notable AMD novelty among the supercomputing kits is the announcement of the Frontier supercomputer at the US Department of Energy’s Oak Ridge National Laboratory, which displaced Japan’s Arm-based Fugaku cluster to No. 1 spot on the Top500 list of the world’s most powerful. publicly known systems.
Top500 updates its list twice a year and announced its latest update on Monday.
In addition to securing the top spot, Frontier is also a big deal for AMD, as it is the world’s first publicly benchmarked exascale supercomputer to achieve a maximum performance of 1.1 exaflops, based on the standard Linpack benchmark used to measure the world’s best systems.
It was only a few years ago that Intel and the DOE proclaimed that the Intel-powered Aurora supercomputer would become the first exascale system in the United States, but delays have pushed the timeline back to sometime later this year. These delays apparently prompted Intel sometime last year to revise a 2019 press release to change the originally stated delivery timeline from 2021 to 2022 and remove the mention of the Aurora as the first US supercomputer in exascale.
As a fun side note, The Register noticed Intel edited its 2019 press release about Aurora to remove the mention of it being "the first exascale supercomputer" and to change the delivery date from 2021 to 2022. You don't often see companies editing old press releases like this. pic.twitter.com/4HqipenMPD
— Semiconductor News by Dylan Martin (@DylanOnChips) May 31, 2022
The important caveat to AMD’s exascale victory is that Frontier is actually not the world’s fastest supercomputer when considering systems that do not have publicly submitted benchmark results. This is because China apparently does not have one but two systems that have reached a peak performance of 1.3 exaflops, but the system operators have not yet submitted results to the Top500.
AMD: From 6 to 93 supercomputers in five years
When AMD launched its first generation of Epyc chips in 2017, the company’s CPUs accounted for only six of the world’s fastest 500 supercomputers.
Now, AMD’s CPUs are in 93 of the top 500, according to the spring 2022 update just released. That’s nearly one-fifth of the list, which is almost double the share AMD had last spring.
Many of these AMD-powered systems are among the fastest on the list, with the chip designer’s CPUs present in five of the top 10, 10 of the top 20, 26 of the top 50 and 41 of the top 100.
In contrast, Intel’s CPU share of the Top500 has dropped to 388 systems from 464 five years ago, with the list’s spring 2022 update bringing the x86 giant’s CPU share below four-fifths of total systems for the first time in almost 10 years.
Intel’s representation in the fastest supercomputers in the Top500 has also declined, with the company’s CPUs present in one of the top 10, five of the top 20, 15 of the top 50 and 46 of the top 100.
One of the things that has helped AMD gain ground over the last few years is the fact that its Epyc server CPUs have had higher core numbers than Intel’s Xeon CPUs, which makes Epyc suitable for HPC applications that scale well with kernels.
This is reflected in the latest Top500 list, where AMD Epyc cores account for 27 percent of total cores across all systems. Meanwhile, Intel cores represent 45 percent of the total cores, which makes sense since its CPUs are still present in most systems.
It is important to remember that the HPC world does not consist exclusively of x86 chips. There are still a total of 19 supercomputers with chips not designed by Intel or AMD.
Seven use IBM’s Power chips, which power systems in nos. 4 and 5 and account for 5.5 percent of all cores. Five use Fujitsu’s Arm-compatible A64FX chips, which power Japan’s No. 2 Fugaku system and account for nine percent of all cores.
Four use NEC Vector Engine chips, which represent a very low percentage of cores. The No. 6 Sunway TaihuLight system uses China’s home-grown ShenWei chip, which represents 11.6 percent of all cores. Another in China uses the Hygon Dhyana chip, which uses AMD’s first-generation Zen architecture as part of a joint venture.
AMD finally wins out in Nvidia-dominated accelerator territory
From this Top 500 list use 168 supercomputers in Top500 GPUs. This reflects the fact that although HPC applications have increasingly utilized such components, they are still not used by most of the world’s fastest systems. For those who use accelerators, the 157 is from Nvidia. None of the other accelerator vendors even come close to Nvidia’s share, even AMD, which has only had one system using its GPUs in the last few years.
But AMD finally saw a slight increase in GPU share with Top500’s Spring 2022 update, thanks to seven new systems combining AMD’s third-generation Epyc chips with its new Instinct MI250X GPUs, which are more competitive with Nvidia’s GPUs is than previous products.
Among these seven systems with AMD CPUs and GPUs, Frontier and two other supercomputers are in the top 10 on the list. According to an AMD spokesperson, the Epyc chips used in these systems are codenamed Trento and have optimizations in their I / O array to create cache connection with the MI250X, which basically allows the two processor types to share memory more easily .
AMD clearly has a way to go before it can take any meaningful GPU share from Nvidia, especially given that the total number of systems with accelerators continues to rise, giving Nvidia an opportunity to defend its footprint.
Outside of AMD and Nvidia, there are a few curiosities in the accelerator area of the Top500. Two systems, for example, still use Intel’s discontinued Xeon Phi accelerators. A system in China uses the Matrix-2000 accelerator developed by the country’s National University of Defense Technology. Another in China uses a homemade accelerator that is vaguely called “Deep Computing Processor”.
Then there is Japan, which has two systems that use homemade accelerators. One is called PEZY-SC3, which was developed by the country’s PEZY Computing semiconductor company. The other is called MN-Core, which is developed by Japan’s Preferred Networks.
The latest update from Top500 shows momentum for AMD in several ways, but we should remember that Intel, with a new comeback plan from CEO Pat Gelsinger, is hungry to make amends for the mistakes they have made in recent years and create more competitive chips again.
There is also potential for more systems in the future using Arm-based chips and other alternative architectures, especially given that Europe and countries like China are increasingly looking to design their own. Therefore, AMD should not be complacent. ®