Re: an other provokative question??

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 09/06/07 22:54, Tom Lane wrote:
> "Dann Corbit" <DCorbit(at)connx(dot)com> writes:
>>> Relational database pioneer says technology is obsolete
>>> http://www.computerworld.com/action/article.do?command=3DviewArticleBasic&articleId=3D9034619
>
>> This bit is a hint:
>> "Column-oriented databases -- such as the one built by Stonebraker's
>> latest start-up, Andover, Mass.-based Vertica Systems Inc. -- store data
>> vertically in table columns rather than in successive rows."
>
>> Mr. Stonebraker's company sells column oriented databases. So of course
>> the other methods must be "obsolete".
>
> I don't see anything in there where Stonebraker says that relational DBs
> are obsolete. What he suggests is that column-oriented storage might

Does "column-oriented storage" mean that all of the COLUMN_A values
for all 200 million rows are stored together on adjacent pages?

If so, then doing aggregates (the bread and butter of DW) *would*
seem to be faster. But b-tree leaf that points to "a record" would
need num_cols pointers instead of one pointer. Very messy. And large.

Definitely a niche product.

> beat row-oriented storage for a lot of modern applications. He might be
> right (I'm sure not going to bet against the guy who started Postgres)
> but this has not got anything to do with the concept of a relational
> database. It's an implementation detail --- maybe a pretty fundamental
> one, but in principle you could build a DB either way and no user could
> see a semantic difference.

- --
Ron Johnson, Jr.
Jefferson LA USA

Give a man a fish, and he eats for a day.
Hit him with a fish, and he goes away for good!

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFG4OF4S9HxQb37XmcRAtQeAKCGqjOcdmT6ccrbMy/JDOURjYItSACfVu7/
AEdP1gbDPK/MNwCVlCb1IAg=
=PD28
-----END PGP SIGNATURE-----