Managing XML Data Page 5
Why XML
XML provides a neutral, flexible way of exchanging data among different systems,
applications, and organizations. With XML, data is maintained in an extensible self-
describing format to accommodate ever-evolving business needs. XML documents use
tags to describe the data values they contain, and the nesting of tags to express
hierarchical relationships between the data items. XML can describe very structured data
and enforce the structure through XML schemas, but can also describe semi-structured
data which is prevalent in content-oriented applications.
Service oriented architectures (SOA), enterprise application integration (EAI), enterprise
information integration (EII), Web services, the enterprise message bus (ESB), and
standardization efforts in many industries all rely on XML as an underlying technology
for data exchange.
Organizations as well as entire industries have standardized XML schemas to promote
exchange of data and are evolving those schemas to meet changing business needs. These
efforts include ACORD in the insurance industry, FpML® and FIXML in the financial
industry, RosettaNet in supply chain management, ARTS in the retail business, HL7 in
healthcare, XBRL for business reporting, and DITA for authoring, managing, and
publishing documentation in print and on the Web.
Such industry-specific initiatives, as well as regulatory requirements, are driving the
deployment of XML. As more business transactions are conducted through Web-based
interfaces and electronic forms, government agencies and commercial enterprises bear
greater responsibility for preserving the original order, request, claim, trade, or
submission. XML provides a straightforward means of capturing and maintaining the
data associated with these electronic transactions. Indeed, XML documents frequently
represent transaction records in message-based transaction processing systems.
Pros and cons of XML and relational data
As a self-describing data format, XML allows diverse data (with or without an XML
schema) to be stored in a single document or row without sacrificing the ability to search
or aggregate portions of that data. Applications can evolve their XML schemas without
causing any changes to the underlying database schema. While the flexibility of XML
means that examining and interpreting XML data can consume more CPU and I/O
resources than if the same data was stored in relational form, factors such as schema
complexity might make storing the data in relational form impractical.
With more rigid schema definitions, the relational model requires significantly less
interpretation and allows for more optimized data operations. As such, it can provide
very high performance but might fail to meet application requirements for schema
flexibility. The relational data model is very suitable for applications with stable data
structures and predictable access patterns. XML is often more suitable for applications
with complex and variable data structures, and for combining structured and
unstructured information.
评论0