Accelerating digital audit transformation with GBase 8a
By Du Chao
As a member of the audit department of a city commercial bank, amid our digital audit transformation, we aim to increase digitization and provide objective, real data for decision-making. By integrating and analyzing data from the bank's business and management systems, and on the foundation of fulfilling supervisory duties and strengthening internal controls, we leverage an audit data mart to enable comprehensive analysis across business systems, lines of business, institutions, processes, and inter-bank data — delivering more audit solutions for overall operations and management.
1. Greater Confidence in the GBase 8a MPP Cluster Database
After attending the GBase 8a MPP Cluster v9.5 training in May, I gained clearer direction and stronger conviction in building the technical architecture for a next-generation audit data mart. This database clusterBy Du Chao
As a member of the audit department of a city commercial bank, amid our digital audit transformation, we aim to increase digitization and provide objective, real data for decision-making. By integrating and analyzing data from the bank's business and management systems, and on the foundation of fulfilling supervisory duties and strengthening internal controls, we leverage an audit data mart to enable comprehensive analysis across business systems, lines of business, institutions, processes, and inter-bank data — delivering more audit solutions for overall operations and management.
1. Greater Confidence in the GBase 8a MPP Cluster Database
After attending the GBase 8a MPP Cluster v9.5 training in May, I gained clearer direction and stronger conviction in building the technical architecture for a next-generation audit data mart. This database cluster features a columnar storage MPP architecture, using a fully parallel MPP + Shared Nothing distributed architecture. It adopts a multi-active Coordinator (Master) node deployment with a two-tier structure of computing nodes, avoiding single-point performance bottlenecks and single points of failure (SPOF). On hardware, it fully supports cost-effective x86-based PC servers. For data management, it supports multi-column hash distribution, massive data distributed compression, efficient data loading, partitioned table management, user-defined functions, full-text search, security auditing, resource limits, and data warehouse analytics. In terms of storage and performance, the cluster can scale to 300 or more nodes, with each node supporting up to 50TB of data. It supports massive data storage and querying with an independent storage engine, highly aligning with digital audit transformation.
2. GBase 8a MPP Provides Robust Technical Support for Auditing
In the context of digital auditing, the database for an audit data mart must support diverse applications with multiple architectures, reflecting the architectural diversity of the big data era while ensuring compatibility among OldSQL, NewSQL, and NoSQL. Through concentrated study of GBase 8a MPP Cluster, I’ve gained deeper knowledge of its capabilities. It is a Shared Nothing, distributed parallel database cluster built on columnar storage, delivering high performance, high availability, and high scalability. It offers a cost-effective, general-purpose computing platform for managing ultra-large-scale data (TB to PB level), widely used to support data warehouse systems, BI systems, and decision support systems. This perfectly matches the requirements of a commercial bank’s audit data mart for high concurrency and frequent data pulls in analytics.
3. New GBase 8a MPP Training Approach Boosts Overall Results
The training format was innovative and highly attentive to participant experience — unlike one-way lectures, it emphasized interaction and communication. After the training, course notes were organized and shared with participants. As a hands-on course, the instructor provided cloud server environments, database installation environments, database clients, and conducted practical demonstrations. Post-class exercises and periodic assessments played a vital role in consolidating learning outcomes and verifying effectiveness. The instructor’s Q&A and detailed explanations gave participants from various knowledge backgrounds an excellent starting point to quickly master, deepen understanding of, and gain direct exposure to the product for future use.
This training was an excellent start. When our department leader learned that our team’s serious approach earned us the second-highest scores, they took great notice. They have since requested that younger department members build up their knowledge of homegrown databases, show support for domestic database solutions, and lay a solid foundation for their use in auditing. They also encouraged everyone to maintain this enthusiasm and continue with intermediate and advanced training, so that we can become an elite team of financial audit data analysts.It features a columnar storage MPP architecture, using a fully parallel MPP + Shared Nothing distributed architecture. It adopts a multi-active Coordinator (Master) node deployment with a two-tier structure of computing nodes, avoiding single-point performance bottlenecks and single points of failure (SPOF). On hardware, it fully supports cost-effective x86-based PC servers. For data management, it supports multi-column hash distribution, massive data distributed compression, efficient data loading, partitioned table management, user-defined functions, full-text search, security auditing, resource limits, and data warehouse analytics. In terms of storage and performance, the cluster can scale to 300 or more nodes, with each node supporting up to 50TB of data. It supports massive data storage and querying with an independent storage engine, highly aligning with digital audit transformation.
2. GBase 8a MPP Provides Robust Technical Support for Auditing
In the context of digital auditing, the database for an audit data mart must support diverse applications with multiple architectures, reflecting the architectural diversity of the big data era while ensuring compatibility among OldSQL, NewSQL, and NoSQL. Through concentrated study of GBase 8a MPP Cluster, I’ve gained deeper knowledge of its capabilities. It is a Shared Nothing, distributed parallel database cluster built on columnar storage, delivering high performance, high availability, and high scalability. It offers a cost-effective, general-purpose computing platform for managing ultra-large-scale data (TB to PB level), widely used to support data warehouse systems, BI systems, and decision support systems. This perfectly matches the requirements of a commercial bank’s audit data mart for high concurrency and frequent data pulls in analytics.
3. New GBase 8a MPP Training Approach Boosts Overall Results
The training format was innovative and highly attentive to participant experience — unlike one-way lectures, it emphasized interaction and communication. After the training, course notes were organized and shared with participants. As a hands-on course, the instructor provided cloud server environments, database installation environments, database clients, and conducted practical demonstrations. Post-class exercises and periodic assessments played a vital role in consolidating learning outcomes and verifying effectiveness. The instructor’s Q&A and detailed explanations gave participants from various knowledge backgrounds an excellent starting point to quickly master, deepen understanding of, and gain direct exposure to the product for future use.
This training was an excellent start. When our department leader learned that our team’s serious approach earned us the second-highest scores, they took great notice. They have since requested that younger department members build up their knowledge of homegrown databases, show support for domestic database solutions, and lay a solid foundation for their use in auditing. They also encouraged everyone to maintain this enthusiasm and continue with intermediate and advanced training, so that we can become an elite team of financial audit data analysts.