Fujitsu’s follow-up to mighty supercomputer chip will power data centers
Fujitsu has announced the next chip following the A64FX, the ARM processor (opens in new tab) which is used in one of the most powerful supercomputers in the world, the Fugaku.
The new chip from the Japanese tech giant was announced in a presentation as part of the Fujitsu ActivateNow: Technology summit (opens in new tab), held at the Computer History Museum in Mountain View, California. Fujitsu CTO Vivek Mahajan said it will be an “Arm-based CPU for Next-gen DC” due sometime in 2028.
With the working title ‘MONAKA’, Fujitsu said The register (opens in new tab) that the focus was on making a high-performance chip with better energy efficiency, and said it wanted to contribute to “the realization of a carbon-neutral and sustainable society”.
Energy savings
The new processor could increase HPC workloads and deliver optimal performance for AI (opens in new tab) and data analytics applications, all while offering “overwhelming energy efficiency” over their peers. Fujitsu claims it will have 1.7 times the application performance with double the performance per watt.
MONAKA is part of the program of the New Energy and Industrial Technology Development Organization (NEDO) research agency in Japan. The goal is to increase energy savings in the country’s data centers by 40% by 2030.
In addition to MONAKA, Fujitsu also contributes to this with other developments, such as low-power accelerators, smart network interface cards (NIC) using photonics, and disaggregation technology.
It is speculated that the new chip will therefore be more similar to those used in servers that focus on cloud storage (opens in new tab)cloud hosting (opens in new tab) and colocation providers (opens in new tab). However, these chips are less about saving energy per se than about making full use of it.
“The next generation DC CPU (MONAKA) we are developing will have a wider range of functions and will prove to be more energy efficient,” a Fujitsu spokesperson told The Register. “The range of potential applications is broader than that of the A64FX, which has special features (e.g. interconnects) specific to Fugaku.”
Details are scarce at this point, it seems likely that MONAKA will have the same standout features as the A64FX, such as 28 Gbps Tofu-D interconnect, high-speed HBM2 stacked memory, and 512-bit Scalable Vector Extensions.