Enterprise Open Source Session
Composite Software Presents: Data Services: A Critical Component of SOA
Data Services: A Critical Component of SOA
Jun. 5, 2006 04:00 PM
IDC predicts that data services - services designed to access data in a well-defined, modular fashion - will exceed the number of process services in a service oriented architecture (SOA). The virtues of SOA are well documented. Benefits include reuse and interoperability of services, enabling IT to quickly and effectively meet business demands. However, these benefits will only be realized if data services are designed and implemented according to SOA principles of lose coupling, standards compliance, vendor neutrality, and modularity. In addition, there are novel considerations, such as protection of back-end data sources, that are not general SOA issues but are critical attributes for data services.
This session will discuss key criteria for effective, efficient, and affordable data services within an SOA, including:
* Support for heterogeneous data sources - data required by data services often resides within enterprise applications or legacy data bases. The data is often stored and accessed in idiosyncratic ways. At minimum this means that (1) the data services architecture must support access to a wide range of data sources, and (2) teams building data services will need a wide variety of skills.
* Support for reuse - creation of multiple data services providing essentially the same information, e.g., a customer's credit standing, is at best wasteful and hard to maintain and at worst a source of subtle errors when different process services receive subtly different results from the different data services. Thus data services must be designed for reuse in a variety of contexts. Ideally projects will maximize reuse by purchasing pre-built data services for common tasks.
* Data source protection - data services bridge SOA into legacy applications and databases. Along with building this bridge comes the responsibility of not allowing unauthorized users to cross the bridge and view unapproved data. In addition, the data services must protect the data sources from query loads beyond what they were designed to handle.
About James CrawfordJames brings to Composite Software 15 years of experience in software research and development in optimization and heuristic analysis. James comes to Composite from Ames Research Center, NASA's center of excellence for information technology, where he was responsible for almost 100 researchers. In stringent NASA-wide competitions in 2005, his teams won five major new awards totaling over $20 million. Prior to joining NASA, James led the Optimization team at i2 Technologies where he led a series of significant Joint Development Projects that gave 12 differentiated product capabilities. Before that, he worked at AT&T Bell Laboratories and co-founded the Computational Intelligence Research Laboratory (CIRL) at the University of Oregon. James has authored over fifteen papers in referred journals and conferences, and holds five patents. James holds PhD and Master's degrees in Computer Science from the University of Texas at Austin, and a BA in Math and Computer Science from Rice University.