Re: [bdbxml] Cross-container query without specifying containers
John Snelson <[email protected]> Tue, 21 Feb 2006 09:17:31 +0000
| Newsgroups | gmane.comp.db.dbxml.general |
|---|---|
| Message-ID | <[email protected]> |
Hi Kiril, DB XML does not keep track of the containers that have been created on your system - that is left for the application to handle. However with an appropriate naming scheme for containers (or some other way), it would be easy to store XML documents in new containers whenever you needed to. With a requirement like yours you really ought to be thinking of periodic XmlContainer replacement (say every month), with archival of the oldest containers at a relevent point. No database (or file system) can handle having data dumped in it continuously, without some kind of scheme like this. Finally, we've not yet had a customer reach the maximum number of documents for a container - probably because other constraints like file system size get in the way first. However, it is entirely possible for us to increase the number of documents to 2^64 if that occurs. John Kiril Stankov wrote: > Thanks for the fast reply! > > The problem is not the size, but the number of XML documents, which can > be stored in a container. > I know that 2^32 is a lot, but I am working on an application, which > needs to keep all previous versions of each item. In terms of few years > the above number of documents may be reached and I want to avoid the > "What we do now?" question. > I suppose that what I need to do is track the number of items and create > a new container when the limit is reached. XQuery then may be used to > iterate over containers (I guess that collection() can take a parameter). > > Regards, > > Kiril. > > > >>> Tom McCubbin <[email protected]> 2/20/2006 8:25:09 pm >>> > In the current version this is not possible. > > And i'm curious, as containers are stored in BDB database files where > Key / Value pairs can be 2^32 ( 4 Gig ) each in size, and the overall db > can be up to 256 terra bytes, of course possibly limited by the platform > / file system, since when does a container cap out @ 4gig? What > platform are your working on? > > -tom > Kiril Stankov wrote: > > > Hi, > > > > I wonder if it is possible to search for documents in all existing > > containers, without explicitly providing containers' names. > > What I try to figure out is what to do when reaching the 2^32 limit of > > a container. Will it be possible to simply create another one and > > start storing data in it, designing all queries in such a way that > > they are container-insensitive. > > > > Thanks in advance! > > > > Kiril. > > -- John Snelson, Berkeley DB XML Engineer Sleepycat Software, Inc http://www.sleepycat.com Contracted to Sleepycat through Parthenon Computing Ltd http://blog.parthcomp.com/dbxml ------------------------------------------ To remove yourself from this list, send an email to [email protected]