Could this issue due to the fact I call createStatement one time and I use this Statement for all the queries in my application ? I ran my application without h2 and dumped all sql request to a file (and no speed the problem in this mode, so the DB is really the bottleneck), so I can reproduce the issue in a basic example more easy to debug (while read line; updateQuery()) maybe this is related to some option like LOG/UNDO LOG/LOCK ? I disabled all of them for the sql dump import and it was very faster.
Le samedi 10 août 2013 17:37:03 UTC+2, Pierre a écrit : > > Hello, > > I'm running h2 in embedded mode, I have 3 tables, each one having severals > millions of rows. The db file size is ~ 2Go. I'm using only one concurrent > connection to the database. Most of the query are INSERT and MERGE and are > very simple (no join). I have a big Heap size (-Xmx10g). > > Sometime it tooks a very long time to execute the query, more than 15 > seconds. The problem is it doesn't affect a specific query it affects all > the queries. > > So approximatly, every 30 seconds a random query will take more than 15 > seconds to execute instead of some milliseconds. And it will lock all my > application because I synchronize on application level (only one connection > to the DB). Beside, more the DB is growing, more this random slow query is > long (it was 13 seconds at the beginning, now 18 seconds). > > Do you have any idea what could be the reason of this ? Does H2 in > embedded mode can handle DB with severals million of rows ? Any tweaks ? > > -- 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/groups/opt_out.
