Re: UNDO and in-place update

On Tue, Nov 22, 2016 at 7:31 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>> - Reading a page that has been recently modified gets significantly
>> more expensive; it is necessary to read the associated UNDO entries
>> and do a bunch of calculation that is significantly more complex than
>> what is required today.

Someone told me that there is something called an interested
transaction list stored in the page header, and from that I infer that
isn't *really* true. I think that unless you're interested in an
affected row, rather than just some row that happens to be on the same
page, you don't really have to worry about it.

> Oracle spends a lot of time on this, and it's really cache-inefficient
> because the data is spread all over. This was what this guy felt in
> circa 2001; I'd have to think that the cache unfriendliness problem is
> much worse for modern hardware.

I imagine that temporal locality helps a lot. Most snapshots will be
interested in the same row version.

--
Peter Geoghegan