|
Designing Canonicals for SOA: Bridging ER and XML Worlds
|
![]() Jeff Pekrul
Data Architect, Business Solutions Engineering
Genentech, Inc.
|
![]() Mehmet Orun
Principal Architect, Enterprise Data Services
Genentech, Inc.
|
|
March 7, 2007
8:30 AM - 9:30 AM
Level: Advanced
If logical data models reflect the business view,
one can think that they would be a strong basis for defining canonical
structures to enable service oriented architecture, application agnostic
archiving, etc. The challenge lies defining structures that can be implemented
in the 'real world', with master data management repositories, inconsistent
reference data across applications, and groups that may not agree with
names and definitions.
This session will be based on actual lessons learnt in seeking to tackle the above, what worked, what did not work, how one can go from an ER model (which is a bidirectional graph) to one or more XML schema structures depending on the technical/business need. Jeff Pekrul is a data architect in the Business Solutions
Engineering group at Genentech, responsible for canonical model development,
and support of data integration projects that use EII technology. Prior
to Genentech, Jeff worked as a data modeler/data architect at Charles
Schwab, Harmony Software and AT&T. Jeff has modeled databases in support
of customer-facing applications in several industries, and has experience
with data warehouse modeling as well. Jeff has an MS in Information
Systems Management, and is a member of the San Francisco chapter of
DAMA.
Mehmet Orun is the Principal Architect for Data Services in the Enterprise Architecture group of Genentech responsible for data architecture and data services infrastructure projects. He was previously a management consultant specializing in integration, migration, and data warehousing projects, and also worked in academic and healthcare IT. His interest and expertise includes data and meta data driven solutions to improve data quality, maintainability, and utility while bridging the gap across applications, processes, and structured and unstructured data sources. Towards this end, Mehmet has earned a double MBA in Management and Management Information Systems, with post graduate studies in Computer Science. Mehmet is the Program Director for the San Francisco chapter of DAMA. |