I'm evaluating sqlite and h2 for embedded readonly mode in a multi-threaded 
environment.

My benchmarking code is at 
https://github.com/fabiankessler/readonly-embedded-dbs-benchmark

My questions for h2:

I'm using ACCESS_MODE_DATA=r
Is there anything that does not work at all, or does not work under certain 
circumstances (like large result set)?
There was an issue with DISTINCT once 
https://groups.google.com/forum/?hl=en#!searchin/h2-database/readonly/h2-database/byZAS6eDCNI/YkOl4oQaTTEJ
With read-only I mean that my original data is not modified in any way, not 
that the medium is read-only. Temporary tables in memory and disk are fine.

I'm using MULTI_THREADED=1. Without that, synchronized h2 code is too much 
of a bottleneck.
Since my jvm app does not need any connection to another read-write h2 db, 
I can afford to set that global setting.
Is there anything I need to be afraid of? The feature has been 
"experimental" for many years.

As a better alternative to MULTI_THREADED, do you still recommend to create 
multiple separate single connections to the same db in read-only 
non-locking mode, and managing these connections in something like a 
connection pool in the code? (I've read that suggestion from Thomas 
somewhere in the forum.)

Fabian

-- 
You received this message because you are subscribed to the Google Groups "H2 
Database" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/h2-database.
For more options, visit https://groups.google.com/d/optout.

Reply via email to