From: Madyshev Egor <E.Madyshev@ftdata.ru>
To: Boris Mironov <boris_mironov@outlook.com>, "pgsql-hackers@postgresql.org"
	<pgsql-hackers@postgresql.org>
Subject: RE: Idea to enhance pgbench by more modes to generate data
 (multi-TXNs, UNNEST, COPY BINARY)
Thread-Topic: Idea to enhance pgbench by more modes to generate data
 (multi-TXNs, UNNEST, COPY BINARY)
Thread-Index: AQHcUwuRPlzLCtdZw06GAdoL15fmHbTwZ0aAgAGpzjGAACRkgIAAEcrGgAQQLgCAAG6d+YAGXmPMgAFzjL6AAV1tS4BpseD8gAEgG5GAAGN+EIAAFWg0gArDj2mAAAeBfoAE2r6A
Date: Mon, 9 Feb 2026 13:19:45 +0000
Message-ID: <d261f79e64ff4a3c84dd4185e3bc9db4@localhost.localdomain>
References: <DS0PR08MB9565D91414C65B3AC363825488CFA@DS0PR08MB9565.namprd08.prod.outlook.com>
 <CAExHW5vtdtd5QnFobxOhbVeh5jk3_61zYRxLqCXEZpO3jOeFDg@mail.gmail.com>
 <DS0PR08MB9565C0B8B45F7B8D78D89EA188CAA@DS0PR08MB9565.namprd08.prod.outlook.com>
 <CAExHW5umH2K5w_5GRCWGrRHfceMjVn72hF=z_A7G+vReryx0uw@mail.gmail.com>
 <DS0PR08MB956560D79EA051E98688F78088CAA@DS0PR08MB9565.namprd08.prod.outlook.com>
 <CAExHW5vFMkBfv9zB3c5gxF=VLkR7dxM2a3pSFsNuz4JYqc8wRA@mail.gmail.com>
 <PH0PR08MB7020CE2AE1B6937BD01B852F88C9A@PH0PR08MB7020.namprd08.prod.outlook.com>
 <PH0PR08MB702059D610C7D84594CD3BB388D5A@PH0PR08MB7020.namprd08.prod.outlook.com>
 <PH0PR08MB70201EC778A20A2BDC8A5F4A88D2A@PH0PR08MB7020.namprd08.prod.outlook.com>
 <PH0PR08MB7020802F14B1AC122AAB6F7588D3A@PH0PR08MB7020.namprd08.prod.outlook.com>
 <PH0PR08MB702075EB12438C2F9A32CD26889EA@PH0PR08MB7020.namprd08.prod.outlook.com>
 <PH0PR08MB70205C9C827602026F60FA07889FA@PH0PR08MB7020.namprd08.prod.outlook.com>
 <c4a589d12584455b9197ef38899297a6@localhost.localdomain>
 <PH0PR08MB70206B42F5A79A518B610815889FA@PH0PR08MB7020.namprd08.prod.outlook.com>
 <PH0PR08MB70204888D724794BA83EA6E58866A@PH0PR08MB7020.namprd08.prod.outlook.com>
 <PH0PR08MB70207B96C493E5777DB4D6428866A@PH0PR08MB7020.namprd08.prod.outlook.com>
In-Reply-To: <PH0PR08MB70207B96C493E5777DB4D6428866A@PH0PR08MB7020.namprd08.prod.outlook.com>
Accept-Language: en-US
Content-Language: ru-RU
Content-Type: text/plain; charset="koi8-r"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Archived-At: <https://www.postgresql.org/message-id/d261f79e64ff4a3c84dd4185e3bc9db4%40localhost.localdomain>
Precedence: bulk

Hello, Boris

Thank you for making new productivity measurements. I`ve reviewed the v7=20
version and have some feedback below.

1. Perhaps we should rename functions initRowMethodBin to=20
initRowMethodBinary, initPopulateTableText to initPopulateTableTextCopy,=20
initPopulateTableBinary to initPopulateTableBinaryCopy,=20
initGenerateDataClientSideText to initGenerateDataClientSideTextFrmt,=20
initGenerateDataClientSideBinary to initGenerateDataClientSideBinaryFrmt=20
for better clarity.

2. The --help output currently describes only the 'M' mode. Should we=20
also add a description for the 'S' mode for completeness?

3. I`m wondering if the default value for the 'filler' column in=20
initCreateTables is necessary? The current functionality seems=20
unaffected, so perhaps we could avoid changing this function to keep the=20
diff minimal.

4. I noticed that vanilla client functionality in M mode is not=20
implemented. Is there a specific reason for this? It seems feasible to=20
implement by passing a counter, similar to how it`s done in=20
initPopulateTableBinary. If there are reasons not to implement it, in=20
mode Mg pgbench should not run mode Sg, it just pg_fatal.

5. In mode client binary format generation c, It would be the right=20
thing to do implement write progress of generating data and 'quiet' mode,=20
as it already implement in client text format generation g.

6. In bufferData, when len =3D=3D 1, we call bufferCharData, which already=
=20
increments bin_copy_buffer_length. However, at the end of bufferCharData=20
we increment it again, leading to a double increment.

7. I suggest adding column count in function initPopulateTableBinary,=20
and initBranchBinary, initTellerBinary, initAccountBinary, to pass it=20
from initAccountBinary to init_row, and use it in functions=20
initBranchBinary, initTellerBinary, initAccountBinary. This will=20
increase the readability of the code, and remove the magic numbers in=20
the addColumnCounter call.

8. I think check and install data_generation_type in function=20
checkInitSteps is not quite right. In the current realization, pgbench=20
allows run data generation many times (dtCdtC...), so i suggest do not=20
touching this functionality. My suggestion would be to revert all the=20
changes from function checkInitSteps, set the data_generation_type in=20
switch in runInitSteps and remove call function checkInitSteps=20
from main.


Best regards,
Egor