Ruby Implementations? about puma

[nono]

FYI, Konstantin Haase has posted http://www.engineyard.com/blog/2012/my-summer-of-open-source/ with some informations about Puma.

[mrrooijen]

Like many other servers, it uses the rapid HTTP parser that comes with Mongrel. It also uses a dynamically sized thread-pool for processing requests in parallel. With Puma, you now have a go to choice when it comes to deploying web applications on Rubinius. And since it does not contain any Rubinius specific code, it also works quite well on JRuby or CRuby.

This is it. Full of win. Thanks! I didn't see that article.

[kyledrake]

I will add a detailed explanation of what Puma provides and update the README via pull request. These are great questions!

[mrrooijen]

@kyledrake thanks!

[evanphx]

Looks like you got it all sorted out! Puma will always work on all ruby implementations, but it being tuned for implementations with full concurrency (Rubinius/JRuby).

As @kyledrake so generously offered, I'll always accept pull requests for doc updates to clarify things!

[mrrooijen]

@evanphx great! Thanks a lot for building it. I suspect this will really push web developers towards more (ram/cpu) efficient deployments using threads, and Rubinius in general. Process-based scaling is just crazy in my opinion, and becomes quite expensive in the long run in various ways. Rather than using REE with CoW friendliness which is stuck in Ruby 1.8.7 performance/features but spawns processes more memory efficient, it just doesn't beat spawning these ultra light threads and having a better GC.

Cheers!