1. 5

Lambda@Edge is now generally available for all customers. You can use this new AWS Lambda feature to run Node.js functions across AWS locations globally without provisioning or managing servers, allowing you to deliver richer, more personalized content with low latency to your customers.


  2. 2

    One side of me thinks “this stuff is great, less ops work, simpler ‘deploys’, possibly even useful design constraints, less hassle all round”. The other thinks “keep your gosh-darn hands off my apps, giant corporate cloud beast”. Can’t help wondering how far we’ll continue along this path before we become entirely passive consumers of the prescribed patterns of 2 or 3 megalithic compute providers. (I mean, insofar as someone creating apps is in any way passive, but you get the point.) All seems rosy right now, I love being able to spin up whatever VMs or containers I want, wherever I want, and I certainly don’t want to manage e.g. a large message bus, or geo-aware DNS myself - but WHERE DOES IT ALL END? Jacked in to the Matrix, that’s where. I’d really love it if stuff worked exactly like this, you just submit functions for events/hooks - or even container-style apps - but they ran on some kind of decentralised computing smart grid. Like a power grid, which anyone could access from any point and submit jobs to, but also which anyone could access from anywhere and submit resources to - whether that be raw processing, function-level task execution, or pubsub/database-style service provision - with addressing just “dealt with”, and some kind of smart scheduling/accounting mechanism built in to the protocols(!) that tallied everything up and sorted everyone’s dealings out. Obviously that doesn’t raise a zillion questions or have any fundamental structural or architectural issues to be dealt with before it can even be considered, and it sure sounds like it could probably use A BLOCKCHAIN for some of that stuff, so I’ll probably knock something up this weekend.