pgbench - extend initialization phase control

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|

pgbench - extend initialization phase control

Fabien COELHO-3

Hello devs,

the attached patch adds some more control on the initialization phase.
In particular, ( and ) allow to begin/commit explicitely, and G generates
the data server-side instead of client side, which might be a good idea
depending on the available bandwidth.

Together with the previously submitted patch about getting stats on the
initialization phase, the idea is to possibly improve this phase, or use
it as a benchmark tool in itself.

--
Fabien.

pgbench-init-extended-1.patch (12K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: pgbench - extend initialization phase control

Ibrar Ahmed-4
Does both client/server side data generation in a single command make sense?
Reply | Threaded
Open this post in threaded view
|

Re: pgbench - extend initialization phase control

Fabien COELHO-3

Hello Ibrar,

> Does both client/server side data generation in a single command make
> sense?

I think yes, especially with the other patch which adds timing measures to
the initialization phases. It really depends what you want to test.

With client-side generation you test the libpq COPY interface and network
protocol for bulk loading.

With server-side generation you are get the final result faster when
network bandwidth is low, and somehow you are testing a different kind of
small query which generates a lot of data.

--
Fabien.


Reply | Threaded
Open this post in threaded view
|

Re: pgbench - extend initialization phase control

Ibrar Ahmed-4
The following review has been posted through the commitfest application:
make installcheck-world:  tested, passed
Implements feature:       tested, passed
Spec compliant:           tested, passed
Documentation:            not tested

Other than that, the patch looks good to me.

The new status of this patch is: Ready for Committer
Reply | Threaded
Open this post in threaded view
|

Re: pgbench - extend initialization phase control

Fabien COELHO-3

Hello Ibrar,

> The following review has been posted through the commitfest application:
> make installcheck-world:  tested, passed
> Implements feature:       tested, passed
> Spec compliant:           tested, passed
> Documentation:            not tested
>
> Other than that, the patch looks good to me.
>
> The new status of this patch is: Ready for Committer

Thanks for the review.

Attached v2 is a rebase after ce8f9467.

--
Fabien.

pgbench-init-extended-2.patch (13K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: pgbench - extend initialization phase control

btendouan
Hello Fabien,

> ---------- Forwarded message ---------
> From: Fabien COELHO <[hidden email]>
> Date: Tue, Jul 16, 2019 at 4:58 PM
> Subject: Re: pgbench - extend initialization phase control
> To: Ibrar Ahmed <[hidden email]>
> Cc: PostgreSQL Developers <[hidden email]>
>
>
>
> Hello Ibrar,
>
>> The following review has been posted through the commitfest
>> application:
>> make installcheck-world:  tested, passed
>> Implements feature:       tested, passed
>> Spec compliant:           tested, passed
>> Documentation:            not tested
>>
>> Other than that, the patch looks good to me.
>>
>> The new status of this patch is: Ready for Committer
>
> Thanks for the review.
>
> Attached v2 is a rebase after ce8f9467.

Thanks for your new patch.

But I failed to apply it. Please rebase it against HEAD.

Regards,

---------
Anna


Reply | Threaded
Open this post in threaded view
|

Re: pgbench - extend initialization phase control

Fabien COELHO-3

>> Attached v2 is a rebase after ce8f9467.

Here is rebase v3.

--
Fabien.

pgbench-init-extended-3.patch (13K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: pgbench - extend initialization phase control

btendouan

>
> Here is rebase v3.

Hi,

Thanks for your new patch.

Failed regression test.
It's necessary to change the first a in “allowed step characters are” to
uppercase A in the regression test of 002_pgbench_no_server.pl.

The behavior of "g" is different between v12 and the patche, and
backward compatibility is lost.
In v12, BEGIN and COMMIT are specified only by choosing "g".
It's a problem that backward compatibility is lost.

When using ( and ) with the -I, the documentation should indicate that
double quotes are required,
and  "v" not be able to enclose in ( and ).

Regards,

--
Anna