Split Queues To Share Cloud Development Environments
One of the reasons mirrord for Teams is a gamechanger for many organizations is that it makes it possible for a whole team to work with one shared cloud environment. They don’t need to run the whole environment for each developer, they don’t need to have their entire architecture running on their laptops, and they don’t need to redeploy to Kubernetes to test new code or debug it. Developers just press “run” in their IDE, and their code is executed remocally : the code runs locally, but with access to any required resources in the Kubernetes cluster or outside of it. With mirrord for Teams, developers can even work on the same service at the same time, and control how to divide between them consumable data like incoming HTTP requests, and now also queue messages.
Remote queue messages can be divided between local processes using a new feature called queue splitting. With queue splitting, developers can configure queue message filters for their mirrord session, and their application will only receive messages that match the filters when it fetches messages from a queue. That way, developers can generate test messages, and know for sure it’s the application running in their IDE that will receive those messages. Additionally, their application will only read messages intended for it, leaving the rest to be read by the remote service (or other developers), and avoiding disruption of the shared environment. The first supported queue service is Amazon Simple Queue Service, with support for Apache Kafka and RabbitMQ coming soon.
