Regional Commercial Bank: Full-Stack Distributed Core System for Financial Service Modernization
Regional Commercial Bank G recognized in recent years that its existing IT systems were increasingly inadequate to support its evolving business and innovation goals. From a technical perspective, the bank urgently needed upgrades in architectural modernization, DevOps toolchain integration, agile delivery, development efficiency, and system migration. From a business standpoint, areas such as precision customer management, data asset utilization, risk detection, regulatory compliance, and operational refinement were all constrained by legacy systems.
As a result, the bank prioritized reconstructing its core system, marking a major shift in its technology roadmap. The new generation of the core banking system is built on a domestically developed distributed database as the foundation for core transactions. Designed with future demand and innovation in mind, the bank implemented full-stack upgrades from hardware and basic software to applications—supporting all business scenarios and laying the groundwork for a robust innovation ecosystem.
As the most critical component of a bank’s IT architecture, rebuilding the core system involved not only transforming the core itself but also revamping a vast number of peripheral systems and business processes. As China’s first full-stack distributed core banking project for city commercial banks, the team tackled major technical challenges through collaborative innovation, trial-and-error iterations, and knowledge accumulation—ultimately creating a nationally replicable benchmark.
The core system was built on an open, distributed architecture using domestic hardware and software, ensuring high reliability and security.
A synchronized cold data store and failover validation mechanism were implemented to ensure disaster recovery and business continuity.
Upstream and downstream system interfaces were adapted using front-end and back-end core conversion platforms, enabling a smooth transition without requiring changes to external systems.
Early-stage prototype validation and full-stack system testing ensured technical feasibility, including multi-core support and stack-wide migration.
• The bank’s daily transaction processing capacity has increased significantly. Under the scenario of testing with 70 million accounts across 44 servers, average account-related transaction response time is under 230 milliseconds, with TPS exceeding 20,000.
• While meeting transaction response time and concurrency requirements, 7×24 stability testing achieved a success rate of 99.999853%.
• End-of-day batch processing performance has improved significantly, with regular processing completed in under 50 minutes and final settlement completed in under 120 minutes.
中电金信鲸Bot RPA是一款面向金融行业客户的机器人流程自动化的开发平台。
-
鲸Bot设计器
鲸Bot设计器集成了浏览器自动化组件,office办公自动化组件,数据库组件,文件处理组件、邮件组件等多种类型组件,通过全栈自动识别技术,自动识别目标元素,使开发过程所见即所得,降低了RPA开发门槛,让业务人员也可以方便地进行业务流程的开发。
-
鲸Bot机器人
通过手动执行、定时执行、控制台远程调用的方式运行自动化流程,配合控制台对外提供API接口,供外部程序调用。通过特有的视觉反馈技术提供统一的异常处理机制,极大降低开发和运维成本,让RPA流程运行更加稳定,而解决金融行业因技术人员产能不足难以支持其业务发展的问题。
Submit your requirements and we will contact you as soon as possible.