Re: Quesion about querying distributed databases - Mailing list pgsql-general

From Ron Johnson
Subject Re: Quesion about querying distributed databases
Date
Msg-id CANzqJaBP6CWiUnAARTS=GJzkk-8sqt-mLZZgz9NOyHs-V4-M5w@mail.gmail.com
Whole thread Raw
In response to Re: Quesion about querying distributed databases  (me nefcanto <sn.1361@gmail.com>)
List pgsql-general
On Wed, Mar 5, 2025 at 9:44 PM me nefcanto <sn.1361@gmail.com> wrote:
I once worked with a monolithic SQL Server database with more than 10 billion records and about 8 Terabytes of data. A single backup took us more than 21 days. It was a nightmare.

25 years ago (meaning much slower hardware), I managed a 1TB database.  Backups took about 4 hours.  Could have gotten it down to two hours if I'd wanted to use more tape drives.

Right now, I manage a 5TB database.  Backups take 110 minutes, and that's when using one channel for all IO, writing to not the fastest NAS, and other 3+TB databases backing up to it at the same time.
 
Almost everybody knows that scaling up has a ceiling

And that ceiling is much, much higher than you think it is.
 
, but scaling out has no boundaries.

Except for complexity and fragility. I bet I could get good scaled up performance out of the amount of hardware you're using to scale out.

--
Death to <Redacted>, and butter sauce.
Don't boil me, I'm still alive.
<Redacted> lobster!

pgsql-general by date:

Previous
From: Rob Sargent
Date:
Subject: Re: Quesion about querying distributed databases
Next
From: me nefcanto
Date:
Subject: Re: Quesion about querying distributed databases