Welcome!
To use the personalized features of this site, please log in or register.
If you have forgotten your username or password, we can help.
My Menu
Saved Items

Statistical Constraints for EAI

John KnapmanContact Information

(7)  IBM, Hursley Park, SO21 2JN Winchester, Hampshire, England
Abstract
Enterprise Application Integration (EAI) often requires a loosely coupled event-based architecture in which timing and statistical measures are important. Statistical constraints are proposed to address some of these requirements, although they may well have broader applicability. This idea is extended to the use of approximate terms such as “some” that can be given a statistical interpretation, as in “Some fault reports give rise to subscriptions.” These can be a convenient way of expressing constraints that are important to the application architecture but are not absolute — abstraction through approximation. They may be used to annotate diagrams in UML and are based on the existing Object Constraint Language (OCL) (1). Some of the examples use a more flexible syntax that is the subject of another paper (2).

Contact Information John Knapman
Email: knapman@uk.ibm.com
Fulltext Preview (Small, Large)
Image of the first page of the fulltext

References secured to subscribers.



Export this chapter
Export this chapter as RIS | Text
 
Remote Address: 38.107.191.109 • Server: mpweb22
HTTP User Agent: CCBot/1.0 (+http://www.commoncrawl.org/bot.html)