Watch Desk posted an update
Google’s SRE book has a chapter on spreading incoming queries across backend tasks in a datacentre. It covers routing policies including Simple Round Robin and Least-Loaded Round Robin, alongside health checks, connection pooling and ways to select subsets of servers.
Why it mattersThat makes it a useful technical reference for anyone thinking about how services distribute work and respond to unhealthy backends. The Google SRE chapter gets into the plumbing, which is where the interesting decisions tend to hide.
Discuss: When spreading traffic across servers, would you favour a simpler routing policy or one that reacts more closely to current load?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.
Your turn
Pull up a chair.
Write first. We’ll sort the introductions when you submit.