I'd be surprised if they didn't leverage something like the SmartOS container capabilities https://www.joyent.com/smartos

Yes

I wrapped libcurl as a port https://github.com/puzza007/katipo. It would be interesting to see how much faster it could be made as a NIF.

I've changed the title to say Riot Games. Hopefully that clarifies for others.

Here's a good description of how Erlang does scheduling

http://jlouisramblings.blogspot.com/2013/01/how-erlang-does-...

Knowing Microsoft they'll come with a keyboard

s/NoSQL/MongoDB/

Yeah, that distinction really bothered me. It's like writing an article about the internet after having used only one website.

tl;dw

I'm not sure that would provide any advantages over Erlang. There's a nice discussion here:

http://erlang.2086793.n4.nabble.com/node-js-compared-to-erla...

There is also the issue of Facebook chat being rolled out in 2008, which I'm assuming Node.js didn't exist by. Node.js would need to provide a significant gain in order to justify refactoring that system.

The Scalaris DHT supports range queries, I think.

http://www.zib.de/CSR/Projects/scalaris/

DHT can certainly support range queries by broadcast and merge. The keyword in the context though is efficiently. Scalaris not a typical Dynamo like DHT. It's based on Chord, so it's possible support range scan with O(logn) metadata RPCs (n is number of nodes.) per query, which I wouldn't call efficient as these RPCs are expensive in that they not very cachable (nature of the DHT, unless the range query is exactly the same.) Typical Bigtable implementation can easily cache portion of metadata with clients and resulting 0 metadata RPC in many cases.