On Wed, Sep 16, 2009 at 11:30 PM, Jordan Ritter <[email protected]> wrote:
>
> Yeah, from a practical standpoint, we measure how long it takes a
> "test run" of 1000 branches to "come online" -- to hit the first
> participant expressions (generate_task).
OK,
thanks for the info.
I've been playing with optimizations at "apply" time for expressions.
I just pushed [1] a tiny commit that spares 1 select per apply, effect
should be visible but not spectacular.
I have a 1 update sparing commit in preparation, its effect is really
visible (concurrent_iterator :branches => 300) :
---8<---
jmettr...@sanma ~/w/ruota (ruote2.0)↑ $ ruby test/bm/ci.rb --dm -c
using engine of class Ruote::Dm::DmPersistedEngine (cache on)
user system total real
run 3.940000 0.190000 4.130000 ( 7.442706)
jmettr...@sanma ~/w/ruota (ruote2.0)↑ $ ruby test/bm/ci.rb --dm -c
using engine of class Ruote::Dm::DmPersistedEngine (cache on)
user system total real
run 2.760000 0.120000 2.880000 ( 4.448846)
--->8---
I can't push it now, 3 tests about error management are still broken.
That will be for tomorrow hopefully.
[1]
http://github.com/jmettraux/ruote/commit/94aaa09535bf0ac15d1727bf51874a3853a12c5d
Best regards,
--
John Mettraux - http://jmettraux.wordpress.com
--~--~---------~--~----~------------~-------~--~----~
you received this message because you are subscribed to the "ruote users" group.
to post : send email to [email protected]
to unsubscribe : send email to [email protected]
more options : http://groups.google.com/group/openwferu-users?hl=en
-~----------~----~----~----~------~----~------~--~---
