[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: DOM vs JDOM
- From: Uche Ogbuji <uche.ogbuji@fourthought.com>
- To: James Robertson <jamesr@steptwo.com.au>
- Date: Thu, 15 Feb 2001 21:04:18 -0700 (MST)
> > > There is nothing that requires a DOM tree to be held in memory other
> > > than the DOM implementation itself. Lazy DOMs allow you to
> > > incrementally retrieve and traverse DOM trees, I believe the Xerces
> > > project was working on such a beast, and the dbXML DOM is partially a
> > > lazy DOM.
> >
> >Yep. 4Suite's DbDom is also a lazy DOM. I agree that the claims quoted
> >from the article are disingenuous.
>
> Is it not true, however, that for applications such
> as XSLT, almost all of the DOM would end up loaded in
> memory?
>
> In my understanding, lazy DOMs only save memory if
> many of the nodes are not accessed at all.
>
> Is this the case?
No. For instance, DbDOM is based on an ODMG object database
implementation. The storage manager implements a caching scheme which
allows nodes to be aged and rotated out.
Cache management is quite a difficult area, but quite established, so I'd
bet most persistent DOMs have such swapping schemes.
--
Uche Ogbuji Principal Consultant
uche.ogbuji@fourthought.com +1 303 583 9900 x 101
Fourthought, Inc. http://Fourthought.com
4735 East Walnut St, Ste. C, Boulder, CO 80301-2537, USA
Software-engineering, knowledge-management, XML, CORBA, Linux, Python