GBase Joins Forces with Kunpeng & Ascend to Build a New Chapter in Third-Generation Intelligent Distributed Databases

On May 9, 2024, GBase, as a key partner of the Kunpeng ecosystem, was invited to participate in the Kunpeng & Ascend Developer Conference 2024 at the Zhongguancun International Innovation Center in Beijing. The conference brought together top global technology experts, business leaders, and industry elites to explore the innovation and development of cutting-edge technologies and share the industry's innovative achievements.

As a leading Chinese database provider, GBase has always upheld the concept of technological innovation, dedicated to providing enterprises with high-performance, high-availability, highly secure, and easily scalable database products and services. Partnering with the Kunpeng ecosystem, it co-creates a reliable foundation for the innovation landscape.

GBase Third-Generation Intelligent Distributed Database

At the openGauss sub-forum ‘Converging Database Innovation Power’, Wang Wei, Product Director of GBase 8c, delivered a keynote speech titled ‘GBase Third-Generation Intelligent Distributed Database Empowering Digital Transformation Across Industries’.

Reviewing the general evolution of distributed databases, the industry has progressed from first-generation sharding/distributed middleware architectures to second-generation native distributed databases, and now to GBase’s latest third-generation intelligent distributed database—GBase 8c. The core features of GBase 8c were detailed. Building on its existing distributed architecture, GBase 8c leverages intelligent machine algorithms and AI capabilities, adding a vector engine and introducing features such as intelligent indexing, intelligent optimization, intelligent O&M, and intelligent security.

The DB intelligence of the third-generation intelligent distributed GBase 8c includes intelligent optimization, intelligent O&M, and intelligent security.

Intelligent Sharding: Leveraging database self-learning to recommend optimal data distribution for users, it reduces migration difficulty, lowers system deployment costs, and shortens time to production. It effectively addresses the issue of reliance on manual experience for distribution key selection during application migration.

Intelligent Cardinality Estimation: Using an in-database Bayesian network model to model the joint distribution of multi-column data samples, it provides more accurate cardinality estimates. This significantly improves the optimizer’s accuracy in choosing plans and operators, thereby enhancing overall database throughput.

Adaptive Plan Selection: Suitable for execution scenarios based on generic cached plans. It explores cached plans through range-based linear expansion and performs plan selection via range coverage matching. This addresses performance issues caused by the inability of traditional single cached plans to adapt to changes in query condition parameters, while also avoiding frequent query optimization invocations.

AI4DB: An autonomous database operation and management framework with monitoring and service-oriented capabilities, replacing repetitive, low-value O&M tasks to reduce the workload of operations personnel.

Intelligent Security: Includes intelligent SQL firewall and intelligent data masking. The SQL firewall uses machine learning to train on users’ SQL queries and establish an expected whitelist, thereby monitoring and mitigating potential attack risks in alert and firewall modes.

In this data-driven era, GBase will continue to refine its full-stack database product portfolio, delivering the best products and services. Together with numerous partners, it will drive innovation and development in China’s database industry, providing strong support for the digital transformation of all industries.