[kepler-users] Modeling client-to-autonomously-updateable-service interractions

[Edward A. Lee]

I believe the best choices would be PN (using its NondeterministicMerge
actor to merge requests) or the newly added "rendezvous" domain (not
in any release yet, but in the CVS repository).

If what you want is a performance model (i.e. a Monte-Carlo simulation
with timing of events explicitly models), then DE would be the best choice.

Edward




At 12:46 PM 1/2/2006 -0600, Maurice Yarrow wrote:
>Content-class: urn:content-classes:message
>Content-Type: multipart/alternative;
>         boundary="----_=_NextPart_001_01C60FCC.C60C0C84"
>
>Hello Kepler/ptolemy users
>
>Which execution domain is best suited to modeling the interactions between 
>n clients and a service which will also be receiving simultaneous updates 
>of its stored information ?
>
>Please let me elaborate just a bit more on this scenario:
>A service exists - presumably a web service or web application - which 
>will be receving updates of some status information but these updates will 
>be on an irregular (indeterminate) basis.  Simultaneously and also on an 
>irregular basis, some number of clients will be contacting this service 
>and requesting this information from the service.
>
>I have downloaded the complete PtolemyII and Kepler sources and compiled 
>them, and have run all the ptolemy and kepler demos in all the various 
>execution domains.  The most interesting of these (for my purposes) 
>appears to be the PN process network domain and the CI component 
>interaction domain.  However, only the CI domain appears to allow a push 
>model of update of actors, and this, coupled with a stack - or in the case 
>of the Router.xml model - a linked-list (used as a FIFO) appear to be 
>applicable to my needs.
>
>But I am interested in any opinions on this subject that may enlighten me 
>and disabuse of my ignorance of model behavior and applicability to the 
>above problem.
>
>Thanks
>Regards
>
>Maurice Yarrow
>Nortel Networks
>_______________________________________________
>Kepler-users mailing list
>Kepler-users at ecoinformatics.org
>http://mercury.nceas.ucsb.edu/ecoinformatics/mailman/listinfo/kepler-users

------------
Edward A. Lee
Professor, Chair of the EE Division, Associate Chair of EECS
231 Cory Hall, UC Berkeley, Berkeley, CA 94720
phone: 510-642-0253 or 510-642-0455, fax: 510-642-2845
eal at eecs.Berkeley.EDU, http://ptolemy.eecs.berkeley.edu/~eal