Re: shared memory stats ideas

Hi,

On 10/20/22 9:17 AM, Drouvot, Bertrand wrote:
> On 10/19/22 8:19 PM, Andres Freund wrote:
>>
>> 2) Split index and table statistics into different types of stats
>>
>> We track both types of statistics in the same format and rename column in
>> views etc to make them somewhat sensible. A number of the "columns" in 
>> index
>> stats are currently unused.
>>
>> If we split the stats for indexes and relations we can have reasonable 
>> names
>> for the fields, shrink the current memory usage by halfing the set of 
>> fields
>> we keep for indexes, and extend the stats in a more targeted fashion.
> 
> I started to work on this.
> I should be able to provide a patch attempt in the next couple of weeks.

Patch submitted and CF entry created: 
https://commitfest.postgresql.org/40/3984/

Regards,

-- 
Bertrand Drouvot
PostgreSQL Contributors Team
RDS Open Source Databases
Amazon Web Services: https://aws.amazon.com