Fujitsu doesn’t have a reputation as a major microprocessor developer the way AMD, Intel, or ARM do, but the company’s supercomputer and HPC work has resulted in multiple generations of high-performance systems, historically built on the SPARC architecture. Now the company is working on the successor to the K supercomputer it built in partnership with RIKEN, the Japanese research institute. And it’s leaving SPARC behind for good in favor of a new, ARM-based architecture.
Where the old K supercomputer was built around the SPARC64 VIIIfx (itself based on SPARC64 VII with a number of HPC modifications), the new machine is a custom implementation of the ARMv8 instruction set. ARM, you may recall, offers the option to license either a specific ARM design like the Cortex-A76 or an architectural license that allows a company to build a CPU of their own design that maintains ARM compatibility. Fujitsu, like AMD, Apple, Intel, and Qualcomm, has chosen to go this route. And the company’s goals for the project are extremely aggressive, with a targeted 100x performance improvement over the K machine by 2021.
If this goal is realized, the post-K supercomputer would be one of the first machines to deliver exascale compute performance and would represent a roughly 10x gain over the current top-ranked supercomputer in the world, the Summit system from IBM and Nvidia.
The new supercomputer will hit these heights in part by using ARM’s new 512-bit extensions for SIMD compute. According to IEEE Spectrum, the new CPUs themselves contain 48 primary CPUs for computation and two assistant cores to handle computational node traffic. There’s also a variant with 48 cores and 4 assistant cores to handle I/O as well as computational nodes, with one physical CPU per node and 384 nodes per rack of space. Memory bandwidth has also reportedly increased by up to an order of magnitude (for competitive reasons, Fujitsu and RIKEN aren’t disclosing how many nodes the new system will be in total). The new computer also adds support for variable precision floating point to allow it to use reduced precision when appropriate, and enhanced double-precision floating point performance compared to the earlier K machine. Nextplatform thinks both the 50-and 52-core CPU variants are prototypes due to currently low yields and that the finished product will likely aim for 64 cores on a 7nm process.
Fujitsu is claiming the machine will only need 30-40MW compared with 12.7MW for the K machine — a startling increase in efficiency based on advanced power management techniques as well as process technology enhancements. The machine will also include “high performance stacked memory” (likely HBM3), and the core as a whole is said to offer “near Xeon-Class performance.” All of this works out to an improvement of 32-42x higher performance per watt compared with the K supercomputer. Part of the improvement will be in process technology, since the K computer was built on 45nm, but there’s clearly a lot of additional work underneath the hood to make that kind of improvement possible.
Pulling off an exascale launch by 2021 would be something of an achievement given that US officials have been publicly wary of such a bet — though it would still keep Horst Simon from having to kick out $2,000 to anyone who took his bet that we wouldn’t have exascale computing in-house by 2020.
MIPS Returns to Silicon Valley, Eyes Burgeoning AI Market
The custom CPU manufacturer MIPS has returned to its roots in Silicon Valley, and it's already plotting its comeback. Can the small CPU firm still find a place at the AI table?
Intel Touts New Quantum Computing Breakthrough, This Time With Silicon
Intel believes it has a path forward for building silicon qubits on conventional foundry technology. If it's right, we might see markedly better scaling and higher performance in the years to come.
Facebook to Build Its Own Silicon for Real-Time Content Filtering
Facebook is reportedly going to start building its own hardware for content monitoring and real-time filtering.