Re: DSSSL side effect-freeness

["Lassi A. Tuura" <Lassi.Tuura@xxxxxxx>]

On Wed, 28 Jan 1998, W. Eliot Kimber wrote:
|> The issue of distributed groves is ultimately one of processing
|> optimization: if you're going to process huge documents in a reasonable
|> amount of time, it's probably unreasonable to construct the hypergrove anew
|> every time you want to process something. Thus, you construct the bits of
|> the grove whenever its source data changes so that the total hypergrove is
|> immediately available when you need it.  Of course, this is at the cost of
|> large amounts of disk storage, but DASD is cheap, right? (Note that I'm
|> assuming that large volumes of data will not be stored as single SGML
|> documents, which is both foolish and impossible. The processing of multiple
|> documents (and non-SGML component objects) always results in a hypergrove.)

Has anybody considered using an object-oriented database (ODBMS) for
these things?  The database we are currently testing (Objectivity/DB,
http://www.objy.com/) supports large amounts of data and large scale
distribution, and it would appear to me that ODBMSes are almost a
perfect solution for this kind of application. It would not take much
wizardy to store the SGML documents in form of objects, and your grove
would essentially be the objects in your database.  DSSSL processing
would then be as simple as getting hold of the grove and processing the
objects you want.  In the case of Objectivity (and maybe some other
distributed ODBMSes), the database will take care of transparent access
to the objects, regardless of where they reside in the world.  As the
database keeps a cache, the penalty of accessing the objects is not that
huge---especially if you are touching objects that are stored close to
each other.

//lat
PS. In the case it matters, by `large amounts of data' I mean tens of
petabytes (peta is next step up from tera, or 10^15; 1PB = million
gigabytes). We are planning on storing all that into one object
database, starting around year 2005 (we'll be practising with terabytes
soon :-).  By `large scale distribution' I mean that the data may end up
being accessed from virtually anywhere in the world. 
--
Lassi.Tuura@xxxxxxx          There's no sunrise without a night



 DSSSList info and archive:  http://www.mulberrytech.com/dsssl/dssslist