Thanks, Glenn, for your insight.  Here's mine...

Glenn wrote:
>
>I don't think [O-R mapping] would not be any easier using
>fine-grained entity beans.

Please explain.

>Coarse grained object graphs allow us more control over how
>we query our RDBMS.

Please explain.

>I would like to hear from Chip from ObjectSpace about atomicity
>in the ODBMS world.  I suspect that a single update of the entire
>object graph would be considered an atomic transaction.

Sunlab's PJama, a transparent serialized-persistence-by-reachability
engine, is fashioned such that "stabilisation of a snapshot is also
the foundation of transactions".  I guess that means fine-grained
transactions won't perform well with large graphs of mostly unmodified
objects.  So persistence, transactions, and graph granularity *are*
inter-related, as a claimed before.  I got the quote from:

http://www.sunlabs.com/research/forest/COM.Sun.Labs.Forest.doc.pos7.abs.html

===========================================================================
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff EJB-INTEREST".  For general help, send email to
[EMAIL PROTECTED] and include in the body of the message "help".

Reply via email to