Re: computing completion tag is expensive for pgbench -S -M prepared

On 2018-06-07 20:34:39 +0100, Simon Riggs wrote:
> On 7 June 2018 at 20:27, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> > Simon Riggs <simon(at)2ndquadrant(dot)com> writes:
> >> If we're going to compress the protocol, it seems sensible to remove
> >> extraneous information first.
> >
> > Breaking the wire protocol was nowhere in this thread.
>
> No, it wasn't. But there is another thread on the subject of
> compressing libpq, which is what I was referring to.
>
> Andres' patch is clearly very efficient at adding the SELECT tag. I am
> questioning if we can remove that need altogether.

That'd be a wire protocol break. We'd have to add compatibilities for
both things in the client, wait a couple years, and then change. Or we
could make it an optional thing based on a client option passed at
connect. Which'd also take a good while. Those seem extremely
disproportionate complicated solutions for the problem. Nor can I
believe that a "SELECT " in the resultset is a meaningful space issue,
making it even worthwhile to break compat in the first place.

Greetings,

Andres Freund