I'll try a profiler later today, but a quick question: the table has a varbinary column. Could the insert..select be converting this to hex and back, as it does in other scenarios?
On Friday, September 13, 2013 6:52:47 AM UTC-7, Brian Craft wrote: > > To eliminate variables I might have overlooked, I wrote the "insert into > ... select" into the loader, so it happens in the same process, right after > the table is built. I get the same result. It's at least an order of > magnitude slower. > > I'm currently on 1.3.171, if that matters. > > > > On Friday, September 13, 2013 5:13:55 AM UTC-7, Brian Craft wrote: >> >> >> >> On Friday, September 13, 2013 4:55:33 AM UTC-7, Noel Grandin wrote: >>> >>> >>> On 2013-09-13 13:53, Brian Craft wrote: >>> >>> >>> Could it be java -client vs -server? I'm running the "insert into ... >>> select" interactively, and I see now that the process has -client. The file >>> loader I'm starting w/o the -client flag, which I think defaults to >>> -server. I'll try changing this. >>> >>> >>> With that performance difference, that would be unlikely. I'm guessing >>> something else is quite different in the two cases. >>> Triggers maybe? Or different DB mode? >>> The H2 performance tool may be some help here. >>> http://h2database.com/html/performance.html#built_in_profiler >>> >> >> No triggers. I'm opening the file as "file;LOG=1;CACHE_SIZE=65536" in >> both cases. Is there some other db mode I should check? >> > -- 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.
