GBase 8a gains recognition for high-speed processing of massive data volumes.

By Chen Huaping

Previously, most systems at Fujian Rural Credit Union ran on Oracle databases. As the business grew and data volumes increased, our data team followed the example of Agricultural Bank of China and gradually migrated the data warehouse from Oracle to GBase, achieving a significant improvement in batch processing performance. This case study can also be found on the GBase official website. During the pandemic last year, I was responsible for a project to build a GBase 8a MPP Cluster database, migrating data processing from Oracle to GBase. For project needs, I enrolled in the GBase 8a V86 training course and passed the certification exam. As the project progressed, I deepened my understanding of GBase 8a. To further consolidate my foundation and learn about the improvements and changes in the new V95 version compared to V86, I signed up for the GBase 8a V95 certification training course this year and studied on my own after work.

Through the two courses, I found the content to be largely similar. They basically cover three aspects: database technical architecture and principles, installation and deployment and management, and database development and tuning. The courses are relatively basic but highly practical and hands-on. They should not be difficult for those with foundational knowledge of other databases, especially MySQL. The data architecture of V86 and V95 is essentially the same—both use a Shared Nothing architecture. However, V95 adds a VC (Virtual Cluster) mechanism, supports multiple hash distribution columns, and has enhanced security encryption. Since I haven't used V95, I still need to explore the other 40-plus new features and enhancements.

In our project, we actually used GBase 8a MPP Cluster V86. We found that GBase delivers high performance for large-scale batch data processing. A batch job on over 10TB of data that previously took Oracle eight to nine hours was completed by GBase in under two hours. Of course, during the migration, we encountered various issues, such as physical server installation and deployment, node scaling, significant differences from Oracle syntax, an enterprise manager tool that was not user-friendly enough, and low concurrent query performance. These were partially resolved or worked around during actual use. Overall, GBase 8a is quite competitive in data analysis applications, and we should leverage its strength in distributed computing performance.

With the implementation of China's IT application innovation initiative, Chinese databases are being further promoted and adopted. GBase, as a key player, already has successful deployments across various industries, and its stability, maintainability, and availability are continuously improving. I sincerely wish GBase continued success and that it truly shines as a beacon of China's database industry.