From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Lutz Fischer <lfischer(at)staffmail(dot)ed(dot)ac(dot)uk> |
Cc: | pgsql-performance(at)postgresql(dot)org |
Subject: | Re: problem with large inserts |
Date: | 2012-12-13 16:10:38 |
Message-ID: | 5810.1355415038@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-performance |
Lutz Fischer <lfischer(at)staffmail(dot)ed(dot)ac(dot)uk> writes:
> I have currently some trouble with inserts into a table
> If I run only [ the select part ]
> it returns 200620 rows in 170649 ms ( thats just under 3 minutes). I
> stopped the actual insert after about 8h.
It should not take 8h to insert 200k rows on any machine made this
century. Frankly, I'm wondering if the insert is doing anything at all,
or is blocked on a lock somewhere. You say there's no concurrent
activity, but how hard did you look? Did you check that, say, the
physical disk file for the table is growing?
> I am running postgresql 9.2 on a windows 2008 R2 server with 256 GB and
> the database is on something like a raid 1+0 (actually a raid1e)
> consisting of 3x4TB disks (limit of what could easily be fitted into the
> server).
A different line of thought is that there's something seriously broken
about the raid configuration. Have you tried basic disk-speed
benchmarks? (Perhaps there's an equivalent of bonnie++ for windows.)
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2012-12-13 16:20:48 | Re: Limit & offset effect on query plans |
Previous Message | Filip Rembiałkowski | 2012-12-13 16:09:19 | Re: problem with large inserts |