abigail at abigail.be
Tue Feb 25 16:37:26 GMT 2014
On Tue, Feb 25, 2014 at 03:59:21PM +0000, Nicholas Clark wrote:
> On Tue, Feb 25, 2014 at 03:25:54PM +0000, Alex Balhatchet wrote:
> > On 25 February 2014 14:59, Aaron Crane <perl at aaroncrane.co.uk> wrote:
> > > That's because, thanks to demerphq++, Perl 5.10 and above have a
> > > built-in trie optimisation which is defeated by the cleverness of the
> > > Regexp::Trie regexes. To take the example from the Regexp::Trie
> > > documentation:
> > Oh my, thanks for the heads up :-) Nice work Perl 5.10!
> This future has been available to one and all for just over six years.
> I believe it's even installed as standard on current RHEL if sysadmins
> are allergic to the idea of building something even more current from
> IIRC booking.com thought that they got roughly a 30% CPU drop in going from
> vendor supplied 5.8.5 to vanilla 5.14.mumble built from source.
> They never did blog this though :-(
Yeah, but from what I heard, most of the gain was in the custom compilation
parameters. I wasn't involved in the process though, so I may be mistaken.
We're moving to 5.18.something very soon -- perhaps someone will blog
More information about the london.pm