Five things we learned from migrating a mission-critical app to the cloud | by National Australia Bank | Apr 2022

Fast Facts — Murex

  • Murex is a global currency (FX) and commodities trading system for NAB operating 24 hours a day in Australia, New Zealand, Asia, UK, Europe and US.
  • It is one of NAB’s top 20 applications and classified as a mission-critical platform and system of registration.
  • It facilitates gross daily FX revenue of tens of billions ($AUD) per day.

Where did we start

  • NO cloud capacity within the NAB Murex application team. We had to expand capacity parallel to delivery.
  • High complexity. Infrastructure, OS and database are all moving to higher versions. Low latency FX trading requirements.
  • Unfamiliar territory. AWS and Murex vendors had no previous experience or credentials comparable to NAB. One of the first mission-critical multi-cloud workloads.
  • Parallel delivery for two clouds. This added complexity to building a portable cloud solution for AWS and Azure.
  • Increased delivery frequency parallel support of a multi-project pipeline. This included critical regulatory programs that required the recording of millions of transactions and associated real-time and batch reporting.

Where did we land?

  • Seamless transition for users with stability and performance from day 1. Compute-intensive batch processing was 33% faster.
  • Solved Capacity and Automation Blockersincreased the cadence of the pipeline of multiple projects through a DevOps transformation to an Amazon Machine Image (AMI)-based approach with scalable compute power.
  • Multi-cloud role model including pre-recording internal compliance and complex stakeholder requirements
  • Behavioral and Process Transformation in Murex to maximize the benefits of the cloud

What we learned

Teach 1: It’s Transformation, Not Lift and Shift

  1. Technical transformation by adapting the Murex platform to cloud native services such as autoscaling groups and maximizing automated interfaces with NAB Enterprise platforms.
  2. Process transformation by reengineering CI/CD automations to an Amazon Machine Image approach that captures all configurations in repositories to ensure repeatability and eliminate configuration variances.
  3. Behavioral Transformation across multiple geographic locations, projects and squadrons enabling delivery flows with rapidly scaling capacity to increase delivery cadence. Our engineers were excited to get closer to becoming full-stack engineers now that they had the infrastructure under their control.

Teach 2: Partners can help you take your abilities from zero to hero

Black and white photo of a cyclist waving to the crowd
Build it, own it, run it and take us from zero to hero

Learn 3: Get to the heart of the matter early, prioritize delivery functions to support

Teach 4: Success Doesn’t Mean Zero Failures

Learn 5: Promote and find champions


Interested in a career in technology at NAB?

About the author

Adrian Aw will speak on this topic at the AWS Community Event in Melbourne, March 2022.


Leave a Reply

Your email address will not be published.

Back to top