Re: [PATCH] expand the units that pg_size_pretty supports on output

On Tue, Jun 15, 2021 at 8:26 AM David Rowley <dgrowleyml(at)gmail(dot)com> wrote:

> On Tue, 15 Jun 2021 at 21:24, <Shinya11(dot)Kato(at)nttdata(dot)com> wrote:
> > Hmmm, I didn't think YB was necessary, but what do others think?
>
> For me personally, without consulting Wikipedia, I know that Petabyte
> comes after Terabyte and then I'm pretty sure it's Exabyte. After
> that, I'd need to check.
>
> Assuming I'm not the only person who can't tell exactly how many bytes
> are in a Yottabyte, would it actually be a readability improvement if
> we started showing these units to people?
>

I hadn't really thought about that TBH; to me it seemed like an
improvement, but I do see that others might not, and adding confusion is
definitely not helpful. That said, it seems like having the code
structured in a way that we can expand via adding an element to a table
instead of the existing way it's written with nested if blocks is still a
useful refactor, whatever we decide the cutoff units should be.

> I'd say there might be some argument to implement as far as PB one
> day, maybe not that far out into the future, especially if we got
> something like built-in clustering. But I just don't think there's any
> need to go all out and take it all the way to YB. There's an above
> zero chance we'll break something of someones by doing this, so I
> think any changes here should be driven off an actual requirement.
>

I got motivated to do this due to some (granted synthetic) work/workloads,
where I was seeing 6+digit TB numbers and thought it was ugly. Looked at
the code and thought the refactor was the way to go, and just stuck all of
the known units in.

> I really think this change is more likely to upset someone than please
> someone.
>

I'd be interested to see reactions from people; to me, it seems a +1, but
seems like -1, 0, +1 all valid opinions here; I'd expect more 0's and +1s,
but I'm probably biased since I wrote this. :-)