My pleasure!

http://basho.com/products/

http://highscalability.com/blog/2014/2/26/the-whatsapp-archi...

Both distributed databases based on erlang (most of which I guess run on erjang anyway) not distributed systems such as boinc.

The OP is talking about the system part

i.e. Not merely using a distributed system such as erlang, couch/mongodb, and pretending you can do distributed systems.

It's funny that you would pick boinc (a rotten mashup of C and PHP with no formal spec) as your exemplar of a distributed system.

There's Riak Core which is a framework for distributed systems in Erlang.

For Common Lisp, ITA Software[0] is a good answer, I would think.

[0] https://en.wikipedia.org/wiki/ITA_Software

I don't know how distributed ITA software's QPX is, but there are some information about how they work around the GC in some cases here (from 2001): http://paulgraham.com/carl.html

Does it mean that ITA services doesn't work in production at Google or that Google does not use any projects written in Common Lisp?

I didn't find any architectural details on QPX after they've been acquired by Google.

For Haskell I remember reading about Cloud Haskell ( https://haskell-distributed.github.io/ ) which allows you to write Erlang-style systems and Sparkle ( https://github.com/tweag/sparkle ) by Tweag ( http://www.tweag.io/ ) which allows you to use Haskell with Apache Spark. But those are tools and libraries and not complete production systems.

I'd love to read about Haskell or Common Lisp distributed system implementations if you have any pointers!

Mnesia? http://erlang.org/doc/man/mnesia.html

Well, for Erlang it's pretty easy. Also including WhatsApp servers, Riak, a lot of router code...

The question is still interesting for the other two languages though.

system. not database. something like boinc for example.

i.e. something doing decent math rather than just reading and writing chunks of data.

since erlang is up to 100 times slower than java for calculation tasks.

thats 100 times more expensive in terms of hardware.

so i doubt its used much for distributed systems.

But i can see the confusion.

Others I've simply never even considered would be chosen.