[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Fwd: Re: XML and JNDI
On Tue, 10 Aug 1999, you wrote:
> >%_I agree that first version of Ozone-XML can be without clustering of DOM nodes,
> but for long term I prefer clustering or paging of DOM nodes (like in IPSI PDOM),
> otherwise traversing of PDOM nodes from the client will cause a network overhead
> (like in Ozone Garage oldtimers example).
I agree. But IPSI PDOM seems to cluster the DOM nodes only on disk. This is
what ozone already does with all objects. The persistence layer of ozone is
based on a cluster engine. Beside this clustering, it may be a good idea to
cluster DOM nodes in one OzoneObject to avoid database internal and database
external communication.
BTW: the single-VM version of ozone is on its way. I should be able to finish
this work next week. Since most of the potential XML servers, that may use
ozone as a DOM repository, have their own API, they will probably use the
single-VM ozone and so network overhead is no problem.
>
> Is there any references/links to Common DOM Query API (CDQA) ?
Vincent's homepage is at http://www.inria.fr/cermics/dbteam/Vincent.Aguilera .
He has not released his work yet. But you can find examples of the X-OQL
frontend there. He expect the first open source release of X-OQL, which
includes CDQA, to be available September. Hopefully I will be able to provide
the pers DOM implementation then, so that we may come up with an integrated
solution - or at least a first view of an integrated solution ;)
I have to stop here. There is a solar eclipse here in germany within the next
hour and of course I have to see that...
Falko
--
______________________________________________________________________
Falko Braeutigam mailto:falko@softwarebuero.de
softwarebuero m&b (SMB) http://www.softwarebuero.de