Sophie

Sophie

distrib > Fedora > 14 > i386 > by-pkgid > 864d1c3c3cd8df4e3a2692faf8776e05 > files > 1147

db4o-doc-7.4-2.fc13.i686.rpm

<html>
  <head>
    <META http-equiv="Content-Type" content="text/html; charset=utf-8">
    <title>OODBMS</title>
    <link rel="stylesheet" type="text/css" href="../../style.css">
  </head>
  <body>
    <div class="CommonContent">
      <div class="CommonContentArea">
        <h1>OODBMS</h1><p>The emergence of distributed data
architectures - in networks, on clients and embedded in "smart"
products such as cars or medical devices - is forcing companies in an array of
industries to look beyond traditional RDBMS technology and ORM for an improved
way to deal with object persistence. They are searching for a solution that can
handle an enormous number of often complex objects, offers powerful replication
and querying capabilities, reduces development and maintainance costs and
requires minimum (zero) administration.&nbsp; </p>

<p>These requirements can be fulfilled by using
an Object Oriented Database Management System (OODBMS). OODBMS provides an
ideal match with object oriented environments like Java and .NET reducing the
cost of development, support and versioning. </p>

<p>Using OODBMS in software projects also better
supports modern Agile software engineering&nbsp;
practices like:</p>

<ul><li>continuous
     refactoring; </li><li>agile
     modeling; </li><li>continuous
     regression testing; </li><li>configuration
     management; </li><li>developer
     "sandboxes".</li></ul>

<p>Object-oriented database technology was first
introduced in the early 1990s with great fanfare. It was wrongly positioned, at
the time, as a replacement for RDBMS technology, and consequently failed to
fulfill that expectation. However, it is now clear that these two database
technologies are complimentary, each with its own strengths and weaknesses, and
thus with its own applicability domain. While, RDBMS technologies are very
successful in large Enterprise environments, where data has static and
relatively flat structure, and where reporting and high volume data analysis
are required, OODBMS are recognized as superior for embedded (invisible to the
end user) applications, where the data has a complex and frequently changing
structure and/or where zero administration is often a requirement. </p>

<p>Additionally, most OODBMS implementations
suffered technical deficiencies that obscured the true strengths inherent in
the technology, such as Native Queries support (i.e. using the syntax and
semantics of popular OO languages for querying), and hence couldn't realize the
true benefits of a uniform object-oriented software development environment for
both the application and for object persistence. The early OODBMS products also
suffered from a lack of an appropriate market focus and business model. </p>

<p>Consequently, nearly all OODBMS market
entrants have since gone through market consolidation and have become focused
on "vertical" target niches, such as defense and healthcare. These
proprietary object-oriented databases tend to be expensive and usually require
heavy vendor support for their complicated, proprietary interfaces. On average,
70% of these companies' revenues are derived from professional services rather
than licensing. Among these are Versant (VSNT, consolidated with POET);
ObjectStore, now the Real-Time division of Progress (PRGS); and Objectivity. As
a result, most IT decision makers have ceased to consider OODBMS technology as
a mainstream option for their deployment projects. </p>

<p>(For more information about OODBMS technology
refer to the <a href="http://www.odbms.org/">ODBMS.ORG website</a>.)</p>



</div>
    </div>
    <div id="footer">
					This revision (1) was last Modified 2007-05-03T10:51:13 by Tetyana.
				</div>
  </body>
</html>