Re: Why is NEXT_EXTENT changing all the time?

> Well now (and hoping that I am reasonable and that the wind is in the west)
>
> I disagree strongly with this suggestion.
>
> For me ALL segments within a tablespace should have an identical extent
> size, and each segment should be allocated to a tablespace with an
> appropriate storage clause. This way a table which wants to grab a next
> extent can always do it (providing the tablespace is not full). In other
> words fragmentation is eliminated.
>
> I utterly fail to see why one would size segments at different sizes, thus
> guaranteeing fragmentation, merely in order to hold one to three years worth
> of data. The only possible 'justification' would be in an environment when
> yearly reviews of database operation were the object. That translates to
> consultancy.

You are absolutely correct. I wasn't dealing with the fact that one should have multiple tablespaces separated by physical device and other tablespaces separated by extent sizes to avoid fragmentation. Just dealing with the thought (shudder) that someone would ever set out to make every NEXT EXTENT in a database the same size.

Daniel A. Morgan Received on Thu Jan 25 2001 - 23:57:43 CST