Daniel asked:

"Can anyone explain what the advantages are of using 
an RDBMS for Zope object storage over the standard file 
system storage?"

The folks I know that have asked for that feature have 
stated their concerns as being:

        Keeping everything in a single large file is risky,
        since if that file becomes corrupted, the entire 
        Web Site will be down until recovery is complete.

                The other side of that one goes something 
                like, "As if we've never seen systems down
                for RDBMS recovery".

        Keeping data in a form that's more accessible by 
        external systems is an advantage.

                And the comeback for that is, "This stuff
                is not tabular, it's Web Content, and Code". 
                It's not like you can run queries against it,
                and print reports.

        Using an RDBMS that has online backup, and quick
        recovery features is cheap insurance.

                Cheap is relative. if you happen to have an
                expensive RDBMS, maybe you can call using it
                for one more applicaiotn "cheap".

                Making a copy of data.fs is pretty easy, and
                recovery is completed by copying over a 
                damaged file. Not exactly rocket science.

                Also, Tranalyzer makes it pretty painless to 
                recover up to the last good update, which is 
                usually more than you can get from a backup!

As you can see, I don't "get" the arguments, but that's what 
people say, and if they happen to be the people who can veto 
Zope, you try to find a way to meet their requirements.

Jerry S.


Zope maillist  -  [EMAIL PROTECTED]
**   No cross posts or HTML encoding!  **
(Related lists - 
 http://lists.zope.org/mailman/listinfo/zope-dev )

Reply via email to