Similarly, there is also a lot of difficult to cross-growing gaps for the XML database (which is the Native-XML database, hereinafter referred to as XMLDB), resulting in so many years of development, but finally has not grown. Reference website: http://www-900.ibm.com/developerWorks/cn/xml/x-data/part5/index.shtml# 1 Narration, about XMLDB now has some questions: (The following is the referenced content)
storage
The storage information in the repository is simple. If you want to store the information already
XML format, then add it directly to the resource library. This may sound good. After all, in the continuous innovation
In the World World, most of the information coming will be embedded
Soap message
XML fragment format. However,
The XML document is broken down and saved to the relational database is not difficult. When you start viewing other features you want to support, this practice will have some benefits. Similarly many of this
One benefit that Native-XML database suppliers advocated is
The Native-XML database can store and query the xenograft document structure. Besides, for structured data issues: Do you really want the structure of information? For use
This advantage of XML documents, when using structured data, it is not a advantage.
Retrieve
I saw it, from
Native-XML database retrieval information seems to be a benefit: the original
XML format retrieves it without any additional encoding, and information can be displayed in a certain style. However, the nature of the structured data retrieval makes this obvious advantage to become a disadvantage. If the information update is huge (for example, receive a single number of megabytes
XML document stock system night update), some
Native-XML platform needs to return to the entire document from the database
- Even if you only share a small part of the document (such as a change process of a particular stock).
other
Native-XML platform will
XML document is saved before the resource library is decomposed, but if there is a complex document structure (just as many structured
XML documents tend to have this structure), this is a bit awkward. In any case, many relational database vendors are currently being thin.
XML serializer wrapper to support the generation of relational data when needed
XML documentation. This makes the programmer easily obtain information that completed the particular task just needed, which has some format, which has a desired pattern, or can be sent to other recognition
XML goal.
search for
search for
Native-XML database has two conventional solution available, which is selected dependent on the database vendor. some
Native-XML databases require which elements or properties are used for indexing, as in the relational database, which are used to index. This information is then used to establish an index so that the search mechanism can be used to quickly locate the matching document. Others when the document is added to the repository
The Native-XML database is to build all the information in the document, which can imagine that this will cause the storage space requirements to rise speed (imagine all the indexes in the relational database!). Search will return a group because these databases are based on documentation.
XML document; then if necessary, the calling program has to process these documents.
Regrettably, this means more complex search, it is very inconvenient. For example, to find out the customer submitting the highest order for a particular part, think that there is a lot of things in the intermediate link. In terms of point to relationships
Native-XML database is not good. As a result, if the data structure is not a pure hierarchy, it is concerned,
Native-XML database is not an appropriate solution. Most Native-XML databases have this powerful feature
- Execute the ability to complete the full-text search, including the entire symbolic word support, the root (all forms of matching a word: now, in the past and during the time) and similar search (
DTD Near XML Schema. In addition, these features are indispensable when using traditional documents, where the context plays an important role in the meaning, and when using structured data, it is far less important.
polymerization
When using relational data, aggregation is one of the most important features they need, in fact it is in the core of online analysis processing (
OLAP).
Native-XML database is particularly poor in performing aggregation tasks. Because the information is either held in the document, it is generally divided into nodes, so it is difficult to collect information and centralize it (seeking, average, etc.) is difficult, in addition, additional code must be added in the intermediate link . If the structured data application requires any analysis processing
- I bet it will need
- Native-XML database will disappoint you.
