Fujitsu Limited has significantly boosted the performance of its Riken Alp Deep Learning Environment (RAIDEN), a computer system for artificial intelligence research the company has originally deployed in 2017 to the RIKEN Center for Advanced Intelligence Project (RIKEN AIP Center), which is the AI research arm of RIKEN.

The system upgrade by Fujitsu was based on an order received from RIKEN, and RIKEN AIP Center aims to put it into operation anytime this month. The upgraded RAIDEN has increased its performance by a considerable margin, moving from an initial total theoretical computational performance of four peta floating point operations per second (PFLOPs) to 54 PFLOPS, placing it in the top tier of Japan's systems.

Since it began operations after system delivery in April 2017, the RIKEN AIP Center in Japan has put RAIDEN for use in research development (R&D) on next generation AI technology. Such cutting-edge AI research is conducted with enormous neural networks, a machine learning method particularly in deep learning.

The increasing scale of neural networks promises to improve factors such as the accuracy with which the networks handle more complex characteristics, but has also led to a drastic increase in computational volume. Moreover, even in AI research beyond deep learning, computational time is increasing due to the increasing complexity of algorithms and the volumes of data involved.

For graphical processing unit (GPU) servers specialized for deep learning, the system is now employing the latest NVIDIA Tesla V100 GPUs from the previous NVIDIA Tesla P100 GPUs. The computational performance of 54 PFLOPS was able to achieve as well because of the increase in the number of DGX-1 servers from 24 to 54.

For the computational cluster servers capable of more general-purpose processing, in addition to the 32 existing Fujitsu Server PRIMERGY RX2530 M2 x86 servers, Fujitsu has also deployed 64 additional Fujitsu Server PRIMERGY CX2550 M4 servers. In addition, it has deployed one Fujitsu Server PRIMERGY RX4770 M4 unit as a compute server that handles high-volume data.