CEP and Agents…
Posted by Paul Vincent
Interesting to see a fascinating blog entry by Roeland Loggen on “Future Patterns of BPM Technology” which showed CEP as a solution to semi-structured processes, between Straight Through Processing for full structured processes and Agent technology and AI for unstructured processes. As it happens, I had just been discussing Agent Technology with agent-guru James Odell, who is leading the OMG standardization efforts around Agent technologies. Earlier this week, reading the description of Event Processing given by Jack Van Hoof in his SOA-EDA blog on “Event Processor Tracks State of Objects”, it had seemed like CEP could easily map to agent technologies . But would it be worthwhile?
- Agent technology handles extreme scaleability issues by going from “database” processing (single instruction, multiple data records) to “via message” processing (instruction decomposition via separate multiple agents and their intercommunication). Note that CEP usually uses a half-way approach: “in-memory and cache”.
.
- Agents are characterised by being autonomous, having interactions, and being adaptive. CEP engines can be autonomous and interactive to the extent that they simply respond to multiple (complex and continuous) events; adaptiveness could be via machine-learning or more commonly via statistical functions. The main attribute of CEP engines is “memory” (of event histories) which presumably is an advantage in “intelligent” distributed agents too!
.
- Agent Communication Languages often combine rules with agent control commands: indeed CEP engines could be considered (or used as) coarse-grained agents (although typically their control and behaviour are relatively fixed compared to some more dynamic agent-based approaches). One suspects that agent communication languages could easily embed the W3C Rule Interchange working group rule format.
.
- One needs to be aware of the latency issues in communications between agents, especially for time-constrained problems. Agent techniques may be ideal for things like continuous scheduling where they might be controlled by a CEP engine that determines when to reschedule.
.
With CEP being associated with event-driven middleware, and such middleware being an integral part of agent communities, it seems there is likely to be some cross-fertilization in future between these technologies.
Further reading
Apart from Jim’s papers, it is interesting to see some real-time / event driven thoughts on agents here, and thoughts on Enterprise Agents here and here.
AI, Agent-based, BAM, Complex Event Processing (CEP), Rules, Trends | 03 April 2008 |
10 Comments
Other Links to this Post
-
Scheduling Agents with Rules Engines « The Complex Event Processing Blog — April 5, 2008 @ 06:34
-
Complex Event Processing (CEP) Blog » DEBS08(2) - Overloaded Agents — July 3, 2008 @ 09:10
RSS feed for comments on this post. TrackBack URI
By PatternStorm, April 4, 2008 @ 03:56
As a I see it CEP engines might evolve to become more of a runtime (i.e. like an application server if you want) into which to deploy and run event processing agents arranged in event processing networks, more than what today actually are (i.e. kind of on-line “database” servers). An industry standard might be required in order for users to be able to define event processing networks in a platform independent way therefore being able to execute them on any compliant event processing runtime (ooups! sorry, engine). The relationship to agent technologies (although I see a stronger relationship to actor models) comes in the way an event processing runtime might be architected: I think it would be a good idea to closely follow actor model and reactive programming principles.
By peter lin, April 5, 2008 @ 07:57
I’m curious. What kind of rule engine does Tibco use? Recently, I wrote up some of my research into extending RETE algorithm for temporal logic. You can find it here
http://jamocha.svn.sourceforge.net/viewvc/jamocha/morendo/doc/Temporal_Logic_Extension_for_RETE_Algorithm.doc?view=log
By vincent, April 15, 2008 @ 02:33
PatternStorm - the main issue with agent-based CEP is the need for history / memory. Do agents share a memory “blackboard”? Or do they somehow become a distributed memory? But for sure, distributed approaches will / are very important…
By vincent, April 15, 2008 @ 02:37
Peter - TIBCO uses a modified Rete algorithm, closely aligned with the OMG PRR standard. I can’t comment on the temporal implementation details, I’m afraid.
By Peter Lin, April 15, 2008 @ 17:12
Hi Paul,
Thanks for responding. I hope my paper is useful to you. In my research, I didn’t find anyone using the approach I describe in my paper.
I also have a Distributed RETE algorithm Said Tabet and I worked on. We filed a patent for it back in 2004. I’ve described the approach on JESS mailing list for anyone that wants an informal description of the approach.
I agree that distributed memory and distributed reasoning will become more important in the future. Once the patent office grants us the patent, i plan to write up a detailed paper explaining the approach. given the huge backlog at the patent office, that’s going to be another 1 or 2.
peter
By vincent, April 16, 2008 @ 06:36
Further discussion also on the CEP Forum at http://forum.complexevents.com/viewtopic.php?f=13&t=75
By Herbert A. Lowe, January 4, 2009 @ 14:22
The combination of using mobile agents along with complex events processing is one area of computer science which requires further research. One very innovative start-up here in the US, Mobile Agent Technologies, is working in this field to combine these software architectural models but incorporate predictive modeling, a rules engine, cloud computing and a knowledge repository as well. Their website is http://www.agentos.net
By vincent, January 5, 2009 @ 01:47
Hi Herb. I’m sure there are a few startups in the space of events - rules - agents / networks - data mining. Their problem is finding an edge over existing solutions: for example, since TIBCO BE3 in 3Q08 we’ve had agent-based rules + queries with a high performance scalable data grid (aka cloud) with the option of Spotfire S+ analytics. Deployed with customers, too. It’s got to be tough to find a niche nowadays…
[PS: are you guys contemplating CEP up in Renton? ]