From: | "Luke Lonergan" <llonergan(at)greenplum(dot)com> |
---|---|
To: | "Worky Workerson" <worky(dot)workerson(at)gmail(dot)com>, "Craig A(dot) James" <cjames(at)modgraph-usa(dot)com> |
Cc: | "Jim C(dot) Nasby" <jim(at)nasby(dot)net>, "Merlin Moncure" <mmoncure(at)gmail(dot)com>, pgsql-performance(at)postgresql(dot)org |
Subject: | Re: Best COPY Performance |
Date: | 2006-10-25 15:06:36 |
Message-ID: | C164CB8C.51BD%llonergan@greenplum.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-performance |
Mr. Worky Workerson,
On 10/25/06 5:03 AM, "Worky Workerson" <worky(dot)workerson(at)gmail(dot)com> wrote:
> However, I am still curious as to the rather slow COPYs from psql to
> local disks. Like I mentioned previously, I was only seeing about 5.7
> MB/s (1.8 GB / 330 seconds), where it seemed like others were doing
> substantially better. What sorts of things should I look into?
It's probable that you have a really poor performing disk configuration.
Judging from earlier results, you may only be getting 3 x 5.7 = 17 MB/s of
write performance to your disks, which is about 1/4 of a single disk drive.
Please run this test and report the time here:
1) Calculate the size of 2x memory in 8KB blocks:
# of blocks = 250,000 x memory_in_GB
Example:
250,000 x 16GB = 4,000,000 blocks
2) Benchmark the time taken to write 2x RAM sequentially to your disk:
time bash -c "dd if=/dev/zero of=bigfile bs=8k count=<# of blocks> &&
sync"
3) Benchmark the time taken to read same:
time dd if=bigfile of=/dev/null bs=8k
- Luke
From | Date | Subject | |
---|---|---|---|
Next Message | Jim C. Nasby | 2006-10-25 15:22:09 | Re: Best COPY Performance |
Previous Message | Merlin Moncure | 2006-10-25 13:20:02 | Re: Best COPY Performance |