Our new .tech series invites guests inside and outside of Form3, discussing developing trends in the development community and shedding light on Form3's own engineering practices.
There are three main options available to run containers on GCP: cloud functions, cloud run and GKE (Kubernetes). Which is the right fit for your workload and what factors should you consider upon making the decision? JB Jean-Baptiste Leroy a customer engineer from GCP and expert on running containers on GCP takes us through the container landscape on GCP. JB informs us on how powerful the three technologies are and how you can mix and match them according to your needs.
18 Dec 2024TRANSCRIPT
EP 56 .tech - Pact contract testing
Micro service architectures are exploding in popularity, but a tough problem that is often overlooked is to ensure all of the services that depend on each other still work together after a change is made. One solution is to deploy them all to a test environment and run end to end tests which results in very slow feedback cycle and keeps the testing far away from the engineer.
Enter Pact, which solves this problem by enabling you to test dependencies between services as part of your local testing. Pact acts as a super charged mock server, enabling you to test your software functions consumes an API in the way the provider expects. Pact also gives you the ability to test that the provider of the pact (server in this case) is actually serving the API in the way that is it contracted. This end to end testing enables your microservices to change with confidence that everything will work together and it brings the feedback cycle very close to the engineer.
In this episode Yousaf Nabi from Smart Bear talks us through how pact solves all of these problems and how you can also use it to solve other dependencies between software such as events when you have an event driven architecture.
13 Nov 2024TRANSCRIPT
Automated Cloud Failure Testing at Form3
A common requirement for a modern technology company is to have a DR strategy and to prove that their DR strategy works. Often this is executed using a manual set of tests run annually. This approach has many pitfalls, the biggest of which being that your tests become out of date the minute a change is made to the system.
In this episode our host Kevin Holditch interviewed Viktor Peacock Head: SRE and Mike Tonks: Staff SRE on how Form3 have approached DR testing with their multi-cloud platform. Listen to the episode to find out from Viktor and Mike how Form3 have automated their DR tests, so they can run them nightly, to catch bad changes before they make it to production.
Inside a recent episode
EP 57 .tech - Running containers on GCP
Published 23 Jan 2025 · Transcript excerpt
[…] So basically that bit of the container, we take it and deploy it, scale it, expose it over internet if you need it to secure it and yeah, that's it. So basically Cloud Run is not built on top, but it supports a canative API. So canative is like a project within the Kubernetes umbrella, the goal of canative is just to provide, let's say, a serverless pod, so just give us a pod specification and then we deploy that. We don't know if there is a community cluster behind the scene, et cetera. So just to make it easy to deploy a container, observe it, scale it, secure it, and yeah, some traffic to that container. […]
Pod Engine is an independent podcast discovery and analytics service and is not affiliated with or endorsed by this podcast. Artwork and show content belong to their owners. Full legal notice.
Explore this show Podcast research with Pod Engine