Re: embedded/"serverless" (Re: serverless postgresql) - Mailing list pgsql-general

From Chris Travers
Subject Re: embedded/"serverless" (Re: serverless postgresql)
Date
Msg-id 003f01c3dbdc$8ab71c10$7f44053d@winxp
Whole thread Raw
In response to serverless postgresql  (Jeff Bowden <jlb@houseofdistraction.com>)
Responses Re: embedded/"serverless" (Re: serverless postgresql)  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-general
From: "David Garamond" <lists@zara.6.isreserved.com>
> I believe the demands for embedded/"serverless" version of PostgreSQL to
> increase significantly once PostgreSQL is natively available on Windows.
> So I would expect that official embedded support to follow quite shortly
> after win32 port has stabilized. :-)

People are always asking for embedded dbms's  without really considering the
consequences.  For example, if you need to share data, you end up with all
the MS Access sorts of issues.  I know because I used to work at Microsoft
in the department that provided support both for Access and the developer
products.  I suspect the support nightmares may be part of the reason for
pushing MSDE, and hence *get away* from the embedded dbms model ;-)

For example, do you realize that the marketing info said that Access
supported 256 *concurrent* users to a database?  At the same time
troubleshooting corruption was nearly unsupported for a while.

I agree with the approach of a wrapper library which would wrap the
startup/shutdown of a postgresql server so that the programmer doesn't have
to worry about the details, but I would add another idea-- namely that the
library should be able to determine whether the server is running remotely,
and simply pass the connection to libpq.  This would also create a
conceptually cleaner framework for configuration of software which may need
to access a local or remote data store.

Best Wishes,
Chris Travers


pgsql-general by date:

Previous
From: "D. Dante Lorenso"
Date:
Subject: User-specific sequences...
Next
From: Tom Lane
Date:
Subject: Re: embedded/"serverless" (Re: serverless postgresql)