Just theoretically assumption. PostgreSQL sometimes may optimise internal
format of data and can sometimes zip data. ASCII data can be zipped better,
then binary random data. Also PostgreSQL sometimes take decision to keep a
column in an external file, if the column is still too big after zip. I
On 3/21/19 6:49 AM, Tom Lane wrote:
"Peter J. Holzer" writes:
On 2019-03-20 13:20:57 +0100, Thomas Güttler wrote:
Strange. I saw a big difference.
What did you test?
I tested inserts.
The graph with the quantiles was for selects.
Hmm, so there are two different code paths being considered
"Peter J. Holzer" writes:
> On 2019-03-20 13:20:57 +0100, Thomas Güttler wrote:
>> Strange. I saw a big difference.
>> What did you test?
>> I tested inserts.
> The graph with the quantiles was for selects.
Hmm, so there are two different code paths being considered here
-- the OP is apparently
On 2019-03-20 13:20:57 +0100, Thomas Güttler wrote:
>
>
> Am 19.03.19 um 20:37 schrieb Peter J. Holzer:
> > On 2019-03-18 15:33:17 +0100, Thomas Güttler wrote:
> > > I did some benchmarking and in my setup there was major
> > > performance difference.
> > >
> > > I tested a ByteA column.
> > >
Am 19.03.19 um 20:37 schrieb Peter J. Holzer:
On 2019-03-18 15:33:17 +0100, Thomas Güttler wrote:
I did some benchmarking and in my setup there was major
performance difference.
I tested a ByteA column.
If I used ascii data the tests took 52 seconds.
If I used random binary data the test to
On 2019-03-18 15:33:17 +0100, Thomas Güttler wrote:
> I did some benchmarking and in my setup there was major
> performance difference.
>
> I tested a ByteA column.
>
> If I used ascii data the tests took 52 seconds.
> If I used random binary data the test took 250 seconds.
>
> binary data is (r
Adrian Klaver writes:
> On 3/18/19 7:33 AM, Thomas Güttler wrote:
>> If I used ascii data the tests took 52 seconds.
>> If I used random binary data the test took 250 seconds.
This doesn't seem terribly surprising in bytea_output = escape
mode. Probably with bytea_output = hex the performance wo
On 3/18/19 7:33 AM, Thomas Güttler wrote:
I did some benchmarking and in my setup there was major
performance difference.
I tested a ByteA column.
What was the test?
If I used ascii data the tests took 52 seconds.
If I used random binary data the test took 250 seconds.
binary data is (roug
