something I would find useful, and I'm sure others would, too...
it'd be great if H2 would auto-detect when I am using a column a lot in
joins (table1.foo_id = table2.bar_id) and I haven't indexed that column,
and index it for me, so that I am not spending days/weeks wondering why
something runs slow...I've now learned to recognize when it's really
slow, and that is always the problem, but find which table/column takes
a while...
would be better if H2 could just say to itself: Hmm, I'm matching
table.column a lot, and it isn't already indexed, I'll just go ahead and
do it. Mostly I encounter this when I've had to add columns later, and
that requires care such that I forget to declare the index too, and with
small test-dbs it's invisible.
-----
I've done it in the middle of big data run, on the fly: saw that H2 was
really slow, gone to the browser console, said "create index on
table.column" and -- blammo! blazing speed increase, cpu drops from 100%
to 0.5% load (linear search string compares are not good)
One is tempted to write one's own sql and always declare all columns on
all tables to be indexed, but of course that means a lot of additional
indexing going on during inserts/updates, which my big app does a lot of.
size-wise I'm only running into the few millions of records per table,
but that will be changing...I've only barely begun to process data :)
-- clint
--
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=en.
