Hej Brian,

Monday, August 9, 2004, 3:24:05 PM, skrev du:

BM> On Aug 9, 2004, at 9:13 AM, Robert S. Sfeir wrote:

>>
>> P.S. Did you post this on Hibernate list???  :-)
>>

BM> gack, I realize this is a joke, but... please don't! The last thing we
BM> need are more O/R pissing matches :-)

Arh - come on - let's have some fun ;)

Nah - seriously, I would imagine that what he had done was to compare
OJB's PersistenceBroker API against Hibernate's session API and
probably had a mapping file without any thought of what should be lazy
and not....

As I remember the PersistenceBroker API don't cache objects during the
session lifetime - thus that by it's nature will have a big impact on
a batch job. In Hibernate stuff is non-lazy per default (element of least surprise)
and thus a naive implemented batch job can be horrifying slow....

...and I do agree with the other poster that OJB can't be 85 times
faster than HB (and vice versa) - it's the usage of the system and the
algorithm used in batchprocessing that does it...not the persistence
layer.

best regards,
[EMAIL PROTECTED]


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to