From: Alvaro Aguayo <aaguayo@opensysperu.com>
To: Firthouse banu <penguinsfairy@gmail.com>
To: pgsql-admin@postgresql.org <pgsql-admin@postgresql.org>
Subject: RE: Bucardo, BDR , pglogical
Date: Tue, 9 Nov 2021 14:08:35 +0000
Message-ID: <SC1P15201MB2238E35D891F4A3E29627ECCAF929@SC1P15201MB2238.LAMP152.PROD.OUTLOOK.COM> (raw)
In-Reply-To: <CAK_Ha4zXQYnBbsvtpWeU234dGV=rtvi5ALkk_eS_KQ_sMiD7=Q@mail.gmail.com>
References: <CAK_Ha4zXQYnBbsvtpWeU234dGV=rtvi5ALkk_eS_KQ_sMiD7=Q@mail.gmail.com>
Hi.
Until a few months ago, I was using BDR for a large database (> 600Gb). Worked pretty well. At the beginning hand some problems with large transactions, but solved them by adapting the program to use smaller transactions; never knew if it was a Postgres problem, a BDR problem, or if it was solved or is still causing trouble somewhere. Decided to stop using it because the release was old, it forced me to stay on PG9.4 and the newer releases were not free software, but commercial.
pglogical is based on the same technology as BDR, but it’s only master-slave. You might find some information about pglogical “multi-master” mode, but it’s just a way to make many masters converge into a single slave.
Never tried Bucardo, but, as far as I know, it’s trigger based, so I would not recommend it for nothing more than testing or small systems, as trigger-based replicators suffer from amplification and thus poor performance on high transaction concurrency.
If you are willing to pay, BDR is probably the better option. If not, you can always change your application to fit a different architecture. You can also consider some PostgreSQL-as-a-service such as the ones offered by Azure, AWS, or GPC, all of them offering high availability.
Regards,
Alvaro Aguayo
Operations Manager
Jr. Prolongación Ayacucho 177 Urb. Santa Eulalia - San Miguel
Telf. (+51-1) 680-5726
E-mail: aaguayo@ocs.pe
From: Firthouse banu <penguinsfairy@gmail.com>
Sent: 09 November 2021 08:22
To: pgsql-admin@postgresql.org
Subject: Bucardo, BDR , pglogical
Best tool to replicate a database to multiple servers and also when user do any changes in replicated database that too should be brought into master database .
Master —— to multiple servers
And in those servers if any changes done then should be bright those to master also . All database should be in sync…..
If anybody have supporting documents please do send me . And I need answers as well.
Thanks
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: pgsql-admin@postgresql.org
Cc: aaguayo@opensysperu.com, penguinsfairy@gmail.com
Subject: RE: Bucardo, BDR , pglogical
In-Reply-To: <SC1P15201MB2238E35D891F4A3E29627ECCAF929@SC1P15201MB2238.LAMP152.PROD.OUTLOOK.COM>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox