Transformation Background

Explosive data growth and increasing system complexity in the fund industry have raised the requirements for IT infrastructure performance, agility, and reliability. In late 2019, Hongde Fund initiated its IT modernization planning. Leveraging SmartX ECP’s proven track record in the industry, the fund decided to conduct a Proof of Concept (POC) to verify its stability and availability.

Validating SmartX ECP Performance and High Availability

During testing, Hongde Fund ran O32 trading system modules—including post-settlement backup and static risk control—on an HCI architecture. The test covered 23 fund products, over 1,000 risk control rules, and 750MB of post-settlement backup data.

Compared to the legacy architecture, which required approximately 15 minutes for a single static risk control module and 25 minutes for both modules, SmartX ECP deployment reduced runtimes to 7 minutes 43 seconds (a 49% reduction) and 11 minutes 7 seconds (a 56% reduction), respectively. Furthermore, in Oracle RAC reliability tests simulating node failure and recovery, the O32 system remained stable with zero service interruption.

SmartX ECP Resource Pool Transitioning from DR to Production to Host Critical O32 and TA Systems

Based on the POC results, Hongde Fund officially deployed SmartX ECP across 9 nodes in phases, including 6 software-defined nodes and 3 appliance-based nodes. Meanwhile, Hongde Fund transitioned the initial DR resource pool into a production environment to host core systems such as O32 and valuation.

Unified Production and DR Pools for Full Asynchronous Replication

Hongde Fund further utilized the asynchronous replication of SMTX OS to back up business applications in the production cluster. This application-level DR solution achieves an RPO of 1 day (customizable) and an RTO of less than 10 minutes (VM boot time), significantly enhancing system availability and reliability.

For the database layer, a log-based replication scheme was implemented to ensure data consistency during failover. By combining this with the VM-level asynchronous replication of business applications, the fund has established a comprehensive and integrated DR framework.

Read more