| Title: | DEC Rdb against the World |
| Moderator: | HERON::GODFRIND |
| Created: | Fri Jun 12 1987 |
| Last Modified: | Thu Feb 23 1995 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 1348 |
| Total number of notes: | 5438 |
We are in a competitive situation againt IBM (AS/400) and Honeywell.
The customer has the impression that creating views on a relational
database increases the database size significantly.
Is the customer wrong? Do our competitors implement views in this
way?
| T.R | Title | User | Personal Name | Date | Lines |
|---|---|---|---|---|---|
| 186.1 | We don't store extra data. | WIBBIN::NOYCE | Bill Noyce, Parallel Processing A/D | Thu Sep 01 1988 16:09 | 12 |
Some of my really old database textbooks say that there are two
kinds of views: those that are pulled together once when defined,
and their data stored in the database, or those that are re-executed
every time a user queries them. Rdb/VMS provides only the second.
I didn't think anyone did the first, but perhaps the Honeywell software
does?? (I suppose you could consider a VAX Data Distributor extraction
to be this first kind of view...)
For IBM S/38, and, I presume, AS/400, "view" may be something a
bit more limited than what Rdb provides, more like an alternate
index to an RMS file. It increases the file size only slightly.
I'm not sure this is the kind of file your customer means, though.
| |||||
| 186.2 | AS/400 views take space | NOVA::BERENSON | VAX Rdb/VMS Veteran | Thu Sep 08 1988 20:14 | 3 |
If the view mechanism on the AS/400 is the same as on the S/38, then each "view" definition results in a new index being created. This not only takes up space, it also slows down updates to the base tables. | |||||