Re: SIMD optimization for list_sort

On Fri, Feb 28, 2025 at 12:43 PM R, Rakshit <rakshit.r@intel.com> wrote:

> > I don't think "another extension might use it someday" makes a very strong case,
> > particularly for something that requires a new dependency.
>
> The x86-simdsort library is an optional dependency in Postgres. Also the new list sort implementation which uses the
x86-simdsortlibrary does not impact any of the existing workflows in Postgres. 

"Optional" and "Does not impact" are not great selling points to get
us to take a 1500 line patch. As we told you in November, list_sort
isn't critical for us. You need to start with the user and work
backwards to the technology. Don't pick a technology and try to sell
people on using it.

> We ran our extension to stress list sort with low cardinality inputs. For eg, for an array of size 100k having
repeatedvalues in the range 1-10 we still see a gain of around 20% in throughput. 
> We will collect more data for low cardinality inputs and with AVX2 too.

Thanks for the news, those are encouraging results.

--
John Naylor
Amazon Web Services