Huawei debuts AI-powered database

Huawei has launched a new version of its GaussDB database that taps artificial intelligence (AI) capabilities to automate and speed up key database management tasks.

The latest iteration of the proprietary database, which forms part of Huawei’s FusionSphere OpenStack cloud platform, is capable of self-tuning, self-diagnosis and self-healing, among other tasks performed while operating and maintaining databases.

Available on-premise, on private clouds and through Huawei’s public cloud service, the database comes with what Huawei has claimed as the “industry’s first reinforcement learning self-tuning algorithm”.

David Wang, Huawei’s executive director of the board and president of ICT strategy and marketing, said the algorithm can improve tuning performance by over 60% in online analytical processing, online transaction processing and hybrid transaction/analytical processing applications.

Wang added that GaussDB now runs on a variety of processor architectures, including x86, ARM, graphical processing unit (GPU) and neural processing unit (NPU), and delivers high-performance data warehousing services to enterprises in the financial, internet, logistics, education and automotive industries.

The support for ARM architecture comes at a time when cheaper and more power-efficient ARM chips are being considered for use in datacentres. Earlier this year, Huawei launched the Kunpeng 920 ARM-based server chip aimed at big data workloads.

“Humanity is entering the age of an intelligent world,” said Wang. “Data is the new factor of production, and intelligence the new productivity. Heterogeneous, intelligent and converged databases will become the key data infrastructure of the financial, government and telecoms industries.”

The GaussDB database, named after German mathematician Carl Friedrich Gauss, follows the launch of Oracle’s Autonomous Database in October 2017. Early customers of Oracle’s cloud-based 18c database, such as Thai payment services company Forth Smart, have reported 80 times faster performance from the software.

Carl Olofson, research vice-president for data management software at IDC, noted that AI and machine learning are increasingly pervasive technologies that are being used in a wide variety of products and services.

While the future of autonomous products is bright, Olofson called for organisations to assess the impact these types of products will have on their IT staff.

Oracle CEO Mark Hurd said in May 2018 that autonomous databases could leave database administrators (DBAs) unemployed. “There are hundreds of thousands of DBAs managing Oracle databases. If all of that moved to the autonomous database, the number would change to zero.”