MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII;
 format=flowed
Content-Transfer-Encoding: 7bit
Date: Sat, 26 Jan 2019 13:45:50 +1100
From: Justin Clift <justin@postgresql.org>
To: Damien Clochard <damien@dalibo.info>
Cc: PostgreSQL Advocacy <pgsql-advocacy@postgresql.org>
Subject: Re: Bye bye Mongo, Hello Postgres
In-Reply-To: <d1b2345eae1895048f819947434e49b3@dalibo.info>
References: <d1b2345eae1895048f819947434e49b3@dalibo.info>
Message-ID: <b4fe9bca623cc0acf5862012a5d1d90d@postgresql.org>
User-Agent: Roundcube Webmail/1.2.3
Precedence: bulk

On 2018-12-20 23:45, Damien Clochard wrote:
> https://www.theguardian.com/info/2018/nov/30/bye-bye-mongo-hello-postgres

Said Hi to the article authors after this hit the list, just to
check how things were going post-migration. :)

Received a reply yesterday (below) from Philip McMahon.  Looks
like PG is working fine as expected. :D

+ Justin

***

Hey Justin

This is a verrry late reply to your message, which got
buried in my inbox over christmas! Thanks very much for
the feedback!

And yes - it's been great! we haven't had any performance
issues since the initial migration. The only problem I
had was trying to find occurrences of a string in an
unindexed field buried 5 levels deep in the JSONB blob
we use for our content item - the query took over 20
minutes before I gave up.

Not that surprising though with 2 million items to scan through.

We're pretty upset that AWS didn't tell us about this
<https://aws.amazon.com/blogs/aws/new-amazon-documentdb-with-mongodb-compatibility-fast-scalable-and-highly-available/>
to be honest - we wouldn't have justified the cost of
migration to PG if we knew that was coming. But ignoring
the cost, we're in a better place now than on some
AWS specific mongo clone thing.

Thanks again for your message and sorry for the sloow reply!

Phil