Re: Design needed: large arrays
Date: 30 Apr 1993 11:11:10 -0400
Message-ID: <1rrfie$7ki_at_access.digex.net>
strobel_at_pine.mcs.gvsu.edu (Steve Strobel) writes:
>Is not an array another term for repeating group, and does relational
>theory have anything to say about this?
An array is not *necessarily* any more a repeating group than a character string is. Think about it: in some languages (e.g., C), a string is explicitly an array of character. The existence of substring extraction functions based on position (i.e., an index value) supports the argument that string-as-array is true in the DBMS world too.
This thread was started by someone who found Oracle DB performance inadequate in storing a 10000x10000 array. That's probably because he's using the wrong tool, i.e., as an implementation of relational theory, Oracle isn't up to the job. What relational theory has to say is "go away, I don't want your business."
The answer is an object DBMS if the data has to be under transaction control, etc., and flat files otherwise. A middle ground is an extended RDBMS like InterBase. I suspect that in the first poster's case, the array data isn't being manipulated under Oracle control, just extracted, so I don't think in this case the necessary extensions constitute a repeating group.
Seth, grimes_at_oecd.fr Received on Fri Apr 30 1993 - 17:11:10 CEST