Re: Merge Join Cartesian
From: Dominic Brooks <dombrooks_at_hotmail.com>
Date: Wed, 10 Jan 2024 21:52:51 +0000
Message-ID: <DBAPR02MB64705240075393605B5B5F60A1692_at_DBAPR02MB6470.eurprd02.prod.outlook.com>
Date: Wed, 10 Jan 2024 21:52:51 +0000
Message-ID: <DBAPR02MB64705240075393605B5B5F60A1692_at_DBAPR02MB6470.eurprd02.prod.outlook.com>
I would imagine it’s stats or at least the optimiser’s estimates of cardinality between the joining datasets and so probably you see e-rows estimates of 1 from the driving rowsource of that merge join operation.
Cheers,
Dominic
Sent from my iPhone
> On 10 Jan 2024, at 21:32, Amit Saroha <eramitsaroha_at_gmail.com> wrote:
>
>
> Hi All,
>
> I've noticed a lot of queries in our environment that use Merge Join Cartesian instead of Hash Join or Nested Loop Joins, which seems unusual to me. Could you please let me know if you've noticed anything similar and what the cause might be?
>
> Thank you in advance for your assistance and support.
>
> Best Regards,
> AMIT
-- http://www.freelists.org/webpage/oracle-lReceived on Wed Jan 10 2024 - 22:52:51 CET