[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference ulysse::rdb_vms_competition

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

447.0. "POSTGRES replaces INGRES?" by MERIDN::MATTHEWS (It's all Atanasoff's fault!) Mon Oct 02 1989 22:12

    
    In reading through an old (1987) evaluation of VAX database products
    put together by a DEC customer, the following quote caught my eye:
    
    "Dr. Stonebraker at UC Berkeley, the principal designer of INGRES,
    has stated that the current version of INGRES is sufficently
    'hacked-up' that he will not be using it for the implementation
    of POSTGRES."
    
    Can anyone verify this statement?  Can we read into this excerpt
    that existing INGRES applications may be in jeopardy when the next
    generation of RTI software appears?
    
    Just thinking out loud...
T.RTitleUserPersonal
Name
DateLines
447.1Still hope for INGRES SMURF::FRIEDMANJoshua FriedmanTue Oct 03 1989 18:2110
1987 is quite awhile ago.  I know that the new version of INGRES (6.0 and
beyond) has a completely different architecture than the previous versions.
The structure has gone to a client -- (multi-)server arch. and all of the
associated tools and front ends have been rewritten from QUEL to (portable) 
SQL.  There are other major architectural modifications as well.  Perhaps 
with all of these changes Stonebraker may feel more comfortable continuing 
to work with RTI's INGRES.

						--Josh
447.2Postgres is a University Research Project!COOKIE::BERENSONI'm the NRAThu Oct 05 1989 22:558
Stonebraker may have been talking about not using University Ingres,
since he pretty much needs something in the public domain to do Postgres.

Anyone who has read the Postgres papers will know that he is doing
things so differently that Postgres internals have little to do with
Ingres (university or commercial).  For example, he is using an
Rdb/ELN-like logging/recovery scheme (and claiming he came up with it,
ha ha) and not the scheme used by Ingres.  This changes most of the internals.