On Nov 17, 9:40 pm, Thomas Mueller <[email protected]>
wrote:
> Hi,
>
> Maybe the problem is that a new optimization for IN(...) is enabled
> now.

I tried. It is still not in par with 117. My tests consistently show
the same numbers:
123 - no page store - no optimizeInList - +10% more time
123 - with page store - no optimizeInList - +14% more time
123 - no page store -  with optimizeInList - +18% more time
123 - with page store -  with optimizeInList - +35% more time

Needless to say, I'm starting with a completely fresh database with
each test.

>
> > reuseSpaceBtreeIndex
>
> I'm sorry about that. For me, the main priority is the page store
> mechanism. I will not have time to work on the old storage mechanism,
> sorry.
>

I think there's a deeper issue here. System properties are, well,
system-wide. The Java class loading mechanism gives me good separation
when it comes to loading my own version of H2 (I can use 117 while PDT
can use 123), however, system properties still affect the entire
system and that can pose a problem. So, in general, I believe in
moving away from system properties, or at least allowing overrides for
all properties (the optimizeInList is another good example).

Thanks,
Zviki

--

You received this message because you are subscribed to the Google Groups "H2 
Database" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/h2-database?hl=.


Reply via email to