For those who didn't see it on slashdot:
http://www.acmqueue.org/modules.php?name=Content&pa=showpage&pid=293
What's interesting is that PostgreSQL is already working on some of
these things. Of note, there's a patch to allow sequential scans to
'piggyback' on top of other sequential scans. See the quote "For
petabyte-scale databases, the only solution may be to run continuous
data scans, with queries piggybacked on top of the scans." on page 4.
There's also been discussion about how to more intelligently cost UDF's,
something also mentioned on page 4.
--
Jim C. Nasby, Database Consultant decibel@decibel.org
Give your computer some brain candy! www.distributed.net Team #1828
Windows: "Where do you want to go today?"
Linux: "Where do you want to go tomorrow?"
FreeBSD: "Are you guys coming, or what?"