In this edition of DOXLON, Oliver Beattie from Monzo, a banking app, talked us through their network architecture, how they have secured it, and the lessons they've learned along the way. Chris Sinjakli from GoCardless discussed the importance of reliability, and dove into a recent feature for zero-downtime patch upgrades. Last but not least, Ed Hargin and Kyrylo Novotarskyi from TransferWise talked about their move from a monolithic codebase and shared infrastructure towards independent infrastructure management that is better focused on helping solve customer problems.
ZERO-DOWNTIME POSTGRES UPGRADES
On the surface, the tech behind a payments API may look like any other startup’s. You'll probably find some Rails apps, a database, and a bunch of stuff off to the sides to glue it together. GoCardless found that it's mostly not the tech that differs, but the approach.
Using their high-availability Postgres cluster as a running example, Chris Sinjakli, SRE at GoCardless, explores how reliability became so important to them, and dives into the most recent feature they built into the cluster: zero-downtime patch upgrades.
SOLVING NETWORK CHALLENGES AT MONZO
On the one hand, Monzo’s backend is composed of hundreds of microservices running in Kubernetes. On the other, banks need to interconnect to hordes of legacy systems, often involving physical kit and leased lines. This presents an interesting networking challenge. Oliver Beattie, Head of Engineering, goes into detail about how Monzo has designed its network architecture, how they secure it, and lots of lessons learned along the way.
SERVICE DISCOVERY AND CONFIGURATION MANAGEMENT AT TRANSFERWISE
One of the things standing in the way of effective and scalable engineering is shared infrastructure. Ed Hargin, Lead of DevOps and Kyrylo Novotarskyi, Software Engineer for Currencies and Banking, focus on the journey TransferWise took from shared to independent infrastructure management, including moving from legacy manual towards automatic instance provisioning, the evolution of service registration and discovery adoption, and the introduction of centralized distributed configuration storage, including secret management techniques, testing, troubleshooting and disaster recovery scenarios.
This content was provided by #DOXLON, a DevOps monthly event in London, UK, put on by Outlyer and LinuxRecruit, a specialist consultancy within the open source industry that's been introducing great engineers to great organizations across the UK since 2011.
If you'd like to speak at a future DOXLON or join future Meetups, please visit the DOXLON Meetup page.