A new supercomputer has unseated longstanding HPC benchmarks. What does this bode for the future of computational science and human progress?
For the first time in history, one and the same supercomputer has become Number One on three ranking systems simultaneously.
In the Top500 rankings, supercomputers are listed by their performance on the LINPACK Benchmark. Since November 2011, the top machine on the list—the K supercomputer, has never been displaced until now. The new machine that beat it is approximately 2.8 times faster, with a LINPACK performance of 415.53 petaflops and a computing efficiency ratio of 80.87%
The HPCG benchmark is a stand-alone test suite that measures the performance of basic operations such as linear equations composed of a dense coefficient matrix. For this benchmark, a high score of 13,400 TFLOPS (teraflops) has been achieved by the new title holder, exceeding the performance of the Number 2 supercomputer in the benchmark (2,925.75 TFLOPS) by approximately 4.6 times.
Finally, there is the Graph500, which focuses on data intensive supercomputer applications which benchmarks of the past have not addressed fully. When a new set of benchmarks was needed in order to guide the design of hardware architectures and software systems intended to support such applications and to help procurement, Graph500 was conceived.
Graph algorithms are a core part of many analytic workloads. Backed by a steering committee of over 50 international HPC experts from academia, industry, and national laboratories, Graph 500 has established a set of large-scale benchmarks for these applications, such as for concurrent search, optimization (single source shortest path), and edge-oriented (maximal independent set). Further, Graph500 addresses five graph-related business areas: Cybersecurity, Medical Informatics, Data Enrichment, Social Networks, and Symbolic Networks.
In the case of today’s fastest contender, it reached a score of 70,980 gigaTEPs, more than doubling the score of 31,303 gigaTEPS attained by the K computer and tripling the performance of the previous No. 2 on the list which scored 23,756 gigaTEPS.
What this achievement bodes for Society 5.0
First, let us reveal the supercomputer that has topped all three highly demanding ranking systems. The new king of benchmarks is the supercomputer Fugaku (an alternate name for Mount Fuji), being developed jointly by Japanese research institution RIKEN and Fujitsu Limited using Arm technology that is also found in Apple i-devices and which is announced to replace Intel processors for Macintosh computers after 2020.
The record achievement indicates the overall high performance of Fugaku and its significant contribution to realize what technologists have coined as Society 5.0: “A human-centered society that balances economic advancement with the resolution of social problems by a system that highly integrates cyberspace and physical space.”
Society 5.0 follows from progressing beyond the hunting society (Society 1.0), agricultural society (Society 2.0), industrial society (Society 3.0), and information society (Society 4.0).
How do supercomputers fit into the movement to achieve 5.0? According to Satoshi Matsuoka, Director, Riken-Center for Computational Science (R-CCS): “In addition to its use as a supercomputer, I hope that the leading-edge IT technology developed for it will contribute to major advances on difficult social challenges such as COVID-19.”
According to Arm’s president Rene Haas, scalable high-performance computing has reached a certain level that “illustrates a dramatic shift in the type of compute that has been traditionally used in these powerful machines, and it is proof of the innovation that can happen with flexible computing solutions driven by a strong ecosystem.” This alludes to highly-complex analysis of problems and trends in cybersecurity, medical Informatics, data enrichment, social networks, and symbolic networks, all of which impact or impede societal progress.
The HPC race continues
As history has shown through the four-minute mile, once an unprecedented but difficult breakthrough is shown to be possible, the floodgates for besting the latest and greatest for High Performance Computing will have been unlocked.
With Fugaku’s achievement today, other supercomputer research stalwarts in countries such as the USA and China will be hot on the heels of even higher compute performance. Quantum computing may also one day reach large scale viability to add new dimensions of knowledge that have defied humans to this day.
Regardless of supercomputer rankings and competitions to increase the stakes, the effective outcome of developing ever-faster computers will hopefully be the advancement of human knowledge, consciousness and unity beyond mere economics and politics.