Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
![]() |
![]() |
Home -> Community -> Usenet -> c.d.o.server -> Re: Clusters physical representation
In article <840jbe$li9$1_at_nnrp1.deja.com>,
markp7832_at_my-deja.com wrote:
> In article <OtL84.113$uG.4165_at_news.siol.net>,
> "Miha ®nidar¹iè" <miha.znidarsic_at_ixtlan-team.si> wrote:
> > Can someone please explain me physical representation of Clusters
> (Cluster
> > organized tables). How and where the PrimaryKey is stored, how are
> other
> > datas? etc
> >
...removed quote from concepts manual and sequential file example...
>
> There are two kinds of clusters, indexed and hash. The cluster index
> can be stored in any tablespace you want to assign it to. Preferably
a
> tablespace devoted to indexes. You can not access the cluster unless
> the cluster index exists. The cluster index is like any other index.
>
........ stuff about hash
>
My statement about the index needs a little more explanation so let us
make that read. The cluster index is like any other index except that
it will store a cluster key value only once since it points to the
block (or first block in a set of chained blocks) that hold all rows
with a cluster key value. Cluster indexes also store nulls keys where
in a normal there are no entries for null keys. Each table in a
cluster may have a primary key that is or is not the cluster index.
Example. Cluster the emp and dept tables on dept_no. The emp table
would have a primary key on emp_no while dept would have its primary
key on the cluster key, dept_no.
--
Mark D. Powell -- The only advice that counts is the advice that
you follow so follow your own advice --
Sent via Deja.com http://www.deja.com/
Before you buy.
Received on Fri Dec 24 1999 - 16:05:26 CST
![]() |
![]() |