Introduction
As my first article, I will write about a topic that in my opinion may cause great changes in the Techologies of Information world.
Everybody knows that Data Bases are an important component for today enterprises because they contain vital information for them in order to work in a suitable way. Besides, it is well known that XML is the latest tool for the data management since it makes it possible to standardize the way computers process information.
Furthermore, XML is causing the appearance of new technologies such as a data base generation. Even though they are involved in an investigation process, they may be in the future an alternative for the Relational databases we have today. This new generation of data bases are based entirely in XML, that's why they're called NATIVE XML DATABASES.
This kind of databases are completely different to the relational ones, wich offer XML support however, they still store the data in a relational format, that is, using tables, records and columns. On the other side, this traditional databases keep the information under the Binary Large Object data type (BLOB). In addition, the main characteristic this databases offer is that users can obtain results for their XML queries. For this reason, this databases belong to a "XML-enabled database" category.
What is a Native XML Database?
There's no accurate definition of a Native XML Database, nevertheless, the DB Initiative for XML Databases define them as " a logic model for XML documents which stores and receives documents following the same patterns ".
What is a document-centered database ?
All the relational databases are centric around data (data-centric databases) because they have atomic data inside their table fields . A Native XML Database has neither table fields nor atomic data. What they store are XML documents, then they are called documents-centric databases.
Caracteristics
There are several products wich provide a variety of Native XML Databases features. Nevertheless, the most common are:
Storage
If we follow our logical deduction, we will infer that Native XML Databases store information on a XML format. However this is just a deduction because this kind of databases have repositories on XML format that may be DOM or Infoset. This repository stores the indexes that are generated with each XML document.
Data Process
The way in which data is processed in this sort of databases is not too advantageous as it seems to be. In fact, the hierarchical format in which data is stored requires that users recuperate the whole XML document, next they have to upgrade it with the most appropriate XML API and finally it's necessary to save the document in the repository. This is because there isn't a standard language that allows upgrading, insertion or elimination of the elements of an XML document. However there's a language called Xupdate (http://www.xmldb.org/xupdate) that permits to upgrade XML documents but it's not standard at all and there's still some database management software that can't work with it.
Search
This kind of database doesn't use SQL as a query language. Instead of that they use Xpath. Some databases make it possible to choose which elements will have an index while others put the index in the hole document content. The difficulty is that complex search can't be made on these databases (for example cross-join ) because Xpath wasn't created to make searches on databases, instead of that they work in a single document. Several databases allow searching using Full-Text Technology, thus the time to get results can be considerably reduced.
Conclusion
I strongly believe that this kind of databases are the most suitable for all the companies that keep their information in several formats. They would allow to convert the data to an XML format using the program you prefer. Consecuently, you can avoid the tiring process of convert one data format into another. Besides, there's no standard language to process information, so this kind of databases are the most appropriate to make searches on stored documents.
If all the dificulties on this databases are overcame, they will become a good alternative for the information storage because of its XML format and moreover the possibility to develop XML based applications such as Web Services.
Resources
XML:DB Initiative for XML Databases: http://www.xmldb.org
XML and Databases: http://www.rpbourret.com/xml/XMLAndDatabases.htm
XML Database Products: http://www.rpbourret.com/xml/XMLDatabaseProds.htm
|
