Architecture and Performance of Royal Bank of Canada’s Online Banking Application
summary
Royal Bank of Canada runs ~40 Cloud Foundry apps (Bluemix local and dedicated) for retail and commercial banking, serving ~6–7 million retail users and ~80M transactions/day. A retail POC rebuilds the UI with AngularJS, uses Trust Association Interceptor, mutual TLS and DataPower, and two Java services (orchestration + backend stub); goal 400–500 TPS sub‑second, actual peak ~900 transactions/sec. Key fixes: remove session affinity, tune Liberty thread pools or override autonomic threading for high mainframe latency, yielding faster releases (years → hours), >90% fewer production incidents, and 40–90% quality gains.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 1:49About RBC
- 2:20Lines of Business
- 14:33Benefits
- 15:15Production Resiliency
- 17:11Summary
- 20:03Questions
related talks
30mBanking on Cloud Foundry Platform: A Case Study of Digital Business Banking Applications
26mReimagining the Banking Core: APIs as the Digital Fabric of Transformation
31mOpen Banking APIs and Architecture at Scale
40mLoosely Coupled - Micro-Frontends And Capital One’s Contact Centers
43mScaling Full-Stack Applications
55mZero Trust Security For AlwaysOn APIs
26mBuilding a sustainable digital bank
22mLarge Scale Angular Applications