The future of Erlang and BEAM

Some time ago I was seriously interested in Erlang (from the world of C ++ / PHP / Java) - and I saw that it was successfully used in the industry, Ericsson , Facebook , Goldman Sachs , etc. So, I thought it would be a great platform for creating high-demand applications, with low latency, with a much cleaner and more understandable language than, for example, Java (for me).

But after the “wow effect” was gone, I found that there are many high-performance Java libraries that seem to solve many problems that Erlang theoretically works best for real-time applications, low latency applications, concurrency, fault tolerance, etc. .d.). Moreover, it seems that there are things that, despite the Erlang profile, it is simply impossible to achieve on BEAM (for example, the parallel LMAX Disruptor structure ).

So the question is: is Erlang still the best platform for building such demanding applications ? Isn't it better if we stick to one very mature (J) virtual machine and try to make it even better than trying to achieve something similar with less resources (OTP team size versus JVM team, supporters, etc.) ? And in general, is it possible to achieve this << 24> using BEAM?

And just to clarify the situation: I do not want to start a fiery war here. I am just curious because I really like Erlang , and I think this is a great platform, and I would like to spend time and effort on creating real projects on it. But I just wanted to know what others can say about it, and if I'm wrong, maybe someone can fix me.

+6
source share

Source: https://habr.com/ru/post/908203/


All Articles