Tom Lane wrote:
>
> mlw <markw@mohawksoft.com> writes:
> > [Index] --> [Target]->[LRU]->[1]->[2]->[3]->[MRU]
>
RE: http://archives.postgresql.org/pgsql-hackers/2002-05/msg00030.php
There are a few variations, but it seems I am making the same assumptions as
Lincln Yeoh. So, you are saying that when a search for a specific tuple
happens, you have to hit every version of the tuple, no matter what? It isn't a
linked list?
I guess I don't understand. Why does it have to visit all of them? If ordering
them from newest tom oldest, and then take the first transaction ID that it
smaller then current transaction id, doesn't that work?