I couldn't read the article (site was down) which is
supposed to claim that RDBMSs "should be considered legacy technology.";
I don't see how the storage stategy necessarily
affects the relationalness of a DBMS. Nice reads anyway:
http://developers.slashdot.org/article.pl?sid=07/09/06/1527258
http://www.databasecolumn.com/2007/09/one-size-fits-all.html
A free columnstore dbms in motion:
http://monetdb.cwi.nl/
Jan Hidders - 07 Sep 2007 11:52 GMT
> I couldn't read the article (site was down) which is
> supposed to claim that RDBMSs "should be considered legacy technology.";
> I don't see how the storage stategy necessarily
> affects the relationalness of a DBMS.
That is not really what Michael Stonebraker is saying. He is only
trying to make the point that it might not necessarily be a good idea
to have one tightly integrated DBMS for all your data storage and data
manpulation needs. Not only because in different situations you might
want very different storage strategies, but also because a more
specialized DBMS might be easier to configure, tune and maintain for
the task at hand. Which data model such DBMSs would have at the
logical level is less relevant for that discussion, although of course
also not completely irrelevant.
> A free columnstore dbms in motion:http://monetdb.cwi.nl/
Yep. Amsterdam rules! ;-)
-- Jan Hidders
David Portas - 07 Sep 2007 21:14 GMT
>I couldn't read the article (site was down) which is
> supposed to claim that RDBMSs "should be considered legacy technology.";
[quoted text clipped - 5 lines]
> A free columnstore dbms in motion:
> http://monetdb.cwi.nl/
The Computerworld headline and the remarks attributed to Stonebraker are
hype. He didn't write about the model, he wrote about the implementations of
it.

Signature
David Portas