- Fujitsu is the latest infrastructure vendor to enter the market of high-performance computing as a service (HPCaaS) with the ‘Fujitsu Computing as a Service (CaaS)’ portfolio.
- Competitors in this market include Hewlett Packard Enterprise (HPE), Lenovo, IBM, Dell Technologies, Atos, Amazon Web Services (AWS), Microsoft Azure, Google Cloud, Oracle Cloud, and Alibaba.
Fujitsu announced the launch of the Fujitsu CaaS portfolio in Japan last month, a services platform to enable commercial organizations to access high-end HPC capabilities to run complex artificial intelligence (AI) workloads via the public cloud. ‘Fujitsu Cloud Service HPC,’ the first services offering made available as part of the CaaS portfolio, is based on Fujitsu’s Supercomputer PRIMEHPC FX1000 servers running on ARM A64X chips, the same processors behind the world’s fastest supercomputer, Fugaku. Fujitsu has combined these supercomputing capabilities with software to deploy a wide range of AI and machine learning (ML) applications.
The company is now taking advanced orders of Fujitsu Cloud Service HPC and will start delivering it to the Japanese market in October 2022, with a worldwide release to follow at a later date. The company is planning to continue adding more services to the CaaS portfolio during the coming months, including its Digital Annealer technology and more AI cloud services. Fujitsu’s Digital Annealer technology simulates quantum computing by utilizing a digital circuit design that mimics quantum phenomena to solve complex combinatorial optimization problems through superposition, quantum tunneling, and entanglement. The architecture enables the free exchange of signals across all bits to facilitate the simultaneous evaluation of a very high number of potential options to quickly deliver insights.
Fujitsu’s CaaS portfolio is the latest addition to an increasingly competitive market to help commercial organizations access HPC via the cloud, and its portfolio is also an important contribution to the Japanese cloud market for HPC and demanding applications such as AI training and inference. Global customers have enjoyed the fruits of the competition among infrastructure vendors (HPE, Dell Technologies, and Lenovo) and hyperscale providers (AWS, Google Cloud, and Microsoft) vying for their share of this market for several years. However, Japanese users wanting to access HPCaaS have been limited to the main hyperscale cloud providers, which are all global in their scope.
The entry of Fujitsu in this market is therefore bound to make an impact. The company is the first infrastructure provider to offer HPCaaS in its native Japan. Moreover, Fujitsu has an outstanding HPC pedigree, having jointly developed – in collaboration with the RIKEN research institute – the Fugaku supercomputer, which currently holds the top spot on the TOP500 list of the most powerful computers in the world. In order to support use cases across many vertical industries, the RIKEN institute is working with Fujitsu to make the CaaS portfolio compatible with Fugaku. The energy efficiency and performance benefits of the ARM A64X processor, combined with Fujitsu’s experience in AI cloud services, means that CaaS is likely to become an important service for connecting the R&D of Fugaku with commercial uses.
The rise of data-intensive technologies such as AI and advanced analytics, which require very powerful computational capabilities, has driven many organizations to adopt HPC for the first time through flexible consumption models. Industries such as finance (for high-frequency trading and fraud prevention), manufacturing (industrial engineering), and healthcare (gene therapy and DNA sequencing) are increasingly using HPC. The CaaS portfolio, included under Fujitsu’s ‘Uvance’ brand, is poised to be a significant contribution to the market.