| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 414.1 | A Couple of Problems | CREDIT::BOOTH | What am I?...An Oracle? | Tue Aug 29 1989 02:35 | 8 |
| A customer I talked to today said he had no luck in bringing up Ingres
(current version)on more than one node in a VAXcluster.
Further, RTI took more than a year to release a version of Ingres that
would support SMP. They still haven't released benchmark results on SMP
systems.
---- Michael Booth
|
| 414.2 | I thought they used VMS lock mgr ? | CSOA1::CARLOTTI | I have an Erie feeling about this... | Tue Aug 29 1989 16:05 | 18 |
| Thanks for the info, Michael.
I'm getting the feeling that the Ingres database must not be too bad,
judging from the fact that there's a ton of competitive info to be found on
Oracle and not much on Ingres. Also, we chose to base our Ultrix database
product on Ingres (after studying several competing products, I would
assume).
Do you have any benchmark data from previous versions of Ingres on a
VAX/VMS platform?
Were there any performance enhancements that we know of in Ingres v6? My
understanding (from an Ingres salesman) was that most of the enhancements
centered around new windowing features.
Thanks again,
Rick C
|
| 414.3 | RTI is just a less visible competitor | COOKIE::BERENSON | VAX Rdb/VMS Veteran | Tue Aug 29 1989 17:55 | 35 |
| There isn't as much competitive dirt on INGRES because we don't lose to
them very often! ORACLE is a much more difficult product to compete
against, primarily because of good marketing not technical competence.
INGRES is perhaps the most technically backwards major RDBMS currently
available. V6 modernized the I/O and Recovery schemes, but left in all
the other restrictions. For example, maximum record size is still
limited to 2008 bytes because they have fixed 2K pages and records can't
span pages. You still can't add new fields to an existing record. Etc.
If I had access to the V6 documentation, I'd probably update the
comparison I did last year.
We chose INGRES to base our ULTRIX product on 40% based on technical
merit and 60% based on business and marketing reasons. Even the 40%
technical merit was NOT based on technical superiority of their product
overall, but based on meeting specific requirements including a number
in the compatibility (with Rdb/VMS) area. Of the other two vendors
seriously considered, one did not understand/meet our compatibility
requirements (even though I believe they would have been a better
partner) and the other was further from ANSI SQL and had severe
marketing and business drawbacks.
INGRES is well established in the UNIX market, and they know how to
build a UNIX RDBMS as well as anyone around. They also come with a
compatible toolkit between ULTRIX and VMS, giving us even more of a
compatibility message.
As for INGRES vs Rdb/VMS on VMS, all of the reasons for choosing Rdb/VMS
over ORACLE apply here as well. There may be technical differences in
how INGRES vs ORACLE compare to Rdb/VMS at any given moment, but the
long term point is that Rdb/VMS is better tuned to VMS (particularly
VAXclusters) and the gap between how Rdb/VMS fits with VMS and how
INGRES/ORACLE fit is going to widen!
Hal
|
| 414.4 | They are NOT supposed to do this | COOKIE::BERENSON | VAX Rdb/VMS Veteran | Tue Aug 29 1989 17:56 | 3 |
| If you really believe that RTI is trying to get the customer to buy
INGRES, after being brought in for an INGRES-TOOLS-FOR-RDB sale, please
contact Jim Steiner.
|
| 414.5 | Leery in Erie! | CSOA1::CARLOTTI | I have an Erie feeling about this... | Wed Aug 30 1989 05:23 | 12 |
| Thanks for the responses, Hal.
Re: .3 - very insightful stuff.
Re: .4 - I can't prove it, it's just that they have some of the techie's
excited about the client/server model and I just don't trust them.
Maybe I'm over-reacting, but I worked hard to get Rdb in there and
I don't feel like fighting the battle all over again.
Thanks again,
Rick C
|