Tom Lane wrote:
> The reason I don't believe there's any fundamental MVCC problem is that
> no such problem showed up in the head-to-head performance tests that
> Great Bridge did about two years ago. GB is now defunct, and I have
> not heard of anyone else willing to stick their neck out far enough to
> publish comparative benchmarks against Oracle. But I still trust the
> results they got.
<irony-mode-on>
You're missing where Mr Peacetree documented how MVCC performance
degraded by 42.37% between versions 7.1 and 7.3.1, as well as his
extensive statistical analysis of the relative behaviours of
PostgreSQL's semantics versus those of DB/2's MVCC implementation.
</irony-mode-off>
> I have helped various people privately with Oracle-to-PG migration
> performance problems, and so far the issues have never been MVCC or
> transaction issues at all. What I've seen is mostly planner
> shortcomings, such as failure to optimize "foo IN (sub-SELECT)"
> decently. Some of these things are already addressed in development
> sources for 7.4.
Ah, but that's just anecdotal evidence...
And if you used radix sorting, that would probably fix it all. (At
least until you discovered that you needed 65 bit addressing to set
sort_mem high enough... Oh, did I neglect to mention anything about
irony?)
--
output = reverse("gro.mca@" "enworbbc")
http://www.ntlug.org/~cbbrowne/oses.html
"Luckily for Microsoft, it's difficult to see a naked emperor in the
dark." --- Ted Lewis, (former) editor-in-chief, IEEE Computer