tag:blogger.com,1999:blog-29701096.post7449795462413328214..comments2023-10-30T08:55:18.553+01:00Comments on SOA and EDA: Is Event Processing revolutionary?Jack van Hoofhttp://www.blogger.com/profile/10073941747649739657noreply@blogger.comBlogger5125tag:blogger.com,1999:blog-29701096.post-36140749291193842092008-11-04T22:25:00.000+01:002008-11-04T22:25:00.000+01:00I know about the Architecture Journal #17.Especial...I know about the Architecture Journal #17.<BR/><BR/>Especially <A HREF="http://msdn.microsoft.com/en-us/library/dd129913.aspx" REL="nofollow">this article</A> is very good.<BR/><BR/>I would recommend a subscription to Architecture Journal to every IT-architect; it's free.Jack van Hoofhttps://www.blogger.com/profile/10073941747649739657noreply@blogger.comtag:blogger.com,1999:blog-29701096.post-16780522801063997572008-11-04T22:10:00.000+01:002008-11-04T22:10:00.000+01:00Hello Jack, just a small note. The latest editiono...Hello Jack, <BR/><BR/>just a small note. The latest <A HREF="http://msdn.microsoft.com/en-us/library/dd129905.aspx" REL="nofollow">edition</A><BR/>of the Microsoft Architecture Journal has a few articles describing the usage of events in distributed systems and in manufactoring environments. They are a nice read.<BR/><BR/> Regards RobAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-29701096.post-11062521575120420702008-11-04T07:35:00.000+01:002008-11-04T07:35:00.000+01:00"Event processing and correlation has evolved from..."Event processing and correlation has evolved from interrupt handling in computer systems and actuator/sensor technologies in industrial processes which already exist for decades."<BR/><BR/>That somewhat understates the difference between simple event-at-a-time event processing (eg of a mouse click, associated callbacks etc), and the continuous processing of events characterized by CEP. In the latter case, the same event can be used in multiple distinct event patterns "at the same time". Tim Bass uses the "sensor data fusion systems" as a historic context for CEP, which is probably a better analogy.<BR/><BR/>Is CEP revolutionary? No, but it is an enabling technology, allowing more widespread BAM-type applications and real-time decisions. IMHO.<BR/><BR/>CheersAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-29701096.post-57595708608927001842008-11-03T21:18:00.000+01:002008-11-03T21:18:00.000+01:00@PatternStorm,Thanks for your contribution. I like...@PatternStorm,<BR/><BR/>Thanks for your contribution. I like your point of view. Well said...<BR/><BR/>-JackJack van Hoofhttps://www.blogger.com/profile/10073941747649739657noreply@blogger.comtag:blogger.com,1999:blog-29701096.post-7960531728710594062008-11-03T21:13:00.000+01:002008-11-03T21:13:00.000+01:00Hi Jack,what might be disruptive, if it ever happe...Hi Jack,<BR/><BR/>what might be disruptive, if it ever happens, is to architect (enterprise) software systems as what they really have become, i.e. reactive systems(*), as opposed to what they were, transformational systems(**). What I mean is that we are architecting what in the end are reactive systems with the old transformational mindset. The role of CEP here is that it improves the good old raective paradigm with interesting new aspects, such as hierarchies of (composite) events and (composite) event detection by pattern matching, that allow us to deal with the complexity of "big" reactive systems, allowing us to get closer to the possibility of architecting enterprise software systems as reactive systems...<BR/><BR/>(*) Reactive System = a system that continuosly interacts with its environment, it never terminates, the environment can start a computation at any time.<BR/><BR/>(**) Transformational System = a system that accepts some input, performs a computation, produces some output and terminates. Accepts input at predetrmined point in time.<BR/><BR/>Regards,<BR/>PatternStormAnonymousnoreply@blogger.com