Posts Tagged: performance


1
Feb 09

Ruby 1.9.1 out

Ruby 1.9.1 is out:

Just a month ago, Antonio Cangiano ran some benchmarks against a Ruby 1.9.1 preview versus several other Ruby interpreters. Ruby 1.9.1 clocked in at at least twice as fast as Ruby 1.8.7. This is great progress and will tone down some of the “but Ruby is slow” type arguments that have gone around, even though in the grand scheme of things it barely makes a dent.

I don’t care much, but doesn’t it bother people that it was so simple to make Ruby twice as fast? It must have been terribly inefficient before. All the wasted CPU cycles with all those sites running Ruby on Rails applications. Energy wasters.



13
Jan 09

Compiling

compilingFor all those compiler constructionists out there: focus on performance. Take any measure to make it happen: separate compilation, parallelization, anything. Right now I’m waiting for an 8 minute compilation the Stratego compiler and it is annoying. Stratego is compiled by the Stratego compiler to C and the C is then compiled by GCC. Did I mention it takes 8 minutes even if I change only 1 line of code? It’s annoying.

And Stratego is actually developed by “us” (i.e. people in our department) so I shouldn’t complain, because hey, if I don’t like it, why don’t I fix it? Probably because I’m too busy complaining about it.

Ok, the compilation just finished. I’m out.