Sponsored Content
Special Forums News, Links, Events and Announcements Complex Event Processing RSS News OMG BPMN2 Conversations vs Event terminology Post 302266193 by Linux Bot on Tuesday 9th of December 2008 05:00:05 PM
Old 12-09-2008
OMG BPMN2 Conversations vs Event terminology

vincent
12-09-2008 08:04 AM
At OMG this week we had an update on the BPMN 2.0 work. BPMN 2.0 is of course primarily for the “human event processing” community, a.k.a BPM, and adds an XML storage mechanism for persisting processes and diagrams as an “alternative” to WfMC’s XPDL. BPMN defines easy-to-read process “orchestrations” or flows - but of course it isn’t for declarative process definitions (for example as lists of rules) nor continuous event processing (as implied for CEP). But it is very widely used.

One new construct in BPMN 2.0 is the (proposed) BPMN Conversation - an aggregation of communications between processes. This is designed to provide a more summary view for users: BPMN conversations describe the “pathways” between processes, as opposed to the individual messages [*1]. One example showed them as process-to-process, process-to-multiple-instances-of-process, and multi-way conversations. Obviously these conversations map to underlying middleware concepts (if the processes are distributed), and also to complex event processing event sources, sinks and agents. Currently the EPTS Glossary defines the term “Event Channels” to describe some aspect of this connection (in TIBCO BusinessEvents, this is the technical messaging system used).

Currently EPTS is starting up the Working Groups to handle things like glossary updates: one suggestion for the EPTS Glossary Working Group might be to extend the Event Channel definition to allow for something like an Event Pathway that represents the same thing as a BPMN 2.0 Conversation. Currently, Event Channel, Pathway and Topic in EPTS are all described as being equivalent. In TIBCO BusinessEvents we model just Channels (message technology) and Destinations (message addresses or topics) - conversations are effectively inferred rather than fixed in a model, but might be an interesting output for a message monitoring tool.

Notes:

Disclosure: TIBCO’s BPM team is involved in the definition of BPMN 2.0.

[1] Analogy: consider the sheep track (indicating the conversation) on the hillside between pasture1 and pasture2 (processes), rather than specific sheep travelling back and forth (message instances representing instances of process transitions).

Image

Source...
 

3 More Discussions You Might Find Interesting

1. IP Networking

newbie terminology question...

is there a difference between a "port" and a "socket"? (2 Replies)
Discussion started by: trostycp
2 Replies

2. UNIX for Dummies Questions & Answers

General Terminology related to the UNIX System

Hello, I am relatively new to the world of programming in general and to the UNIX System in particular. I have a bunch of theoretical questions related to terminology which I have come across in my coursebooks. I hope I post my question at the right place. Any help would be warmly welcome! ... (1 Reply)
Discussion started by: feliks0
1 Replies

3. Debian

hardware printout layout terminology

trying to find online or a book. the words to use to find 'a' or 'the ways and means' to format left margin, right margin etc from man pages to Hardware Printer. while looking to print man pages with a Hardware Printer. I have found lp. and did a man sudoers | lp -p2 to test. The... (1 Reply)
Discussion started by: cowLips
1 Replies
dat_evd_resize(3DAT)				     Direct Access Transport Library Functions				      dat_evd_resize(3DAT)

NAME
dat_evd_resize - modify the size of the event queue of Event Dispatcher SYNOPSIS
cc [ flag... ] file... -ldat [ library... ] #include <dat/udat.h> DAT_RETURN dat_evd_resize( IN DAT_EVD_HANDLE evd_handle, IN DAT_COUNT evd_min_qlen ) PARAMETERS
evd_handle Handle for an instance of Event Dispatcher. evd_min_qlen New number of events the Event Dispatcher event queue must hold. DESCRIPTION
The dat_evd_resize() function modifies the size of the event queue of Event Dispatcher. Resizing of Event Dispatcher event queue should not cause any incoming or current events on the event queue to be lost. If the number of entries on the event queue is larger then the requested evd_min_qlen, the operation can return DAT_INVALID_STATE and not change an instance of Event Dispatcher RETURN VALUES
DAT_SUCCESS The operation was successful. DAT_INVALID_HANDLE The evd_handle parameter is invalid. DAT_INVALID_PARAMETER The evd_min_qlen parameter is invalid DAT_INSUFFICIENT_RESOURCES The operation failed due to resource limitations DAT_INVALID_STATE Invalid parameter. The number of entries on the event queue of the Event Dispatcher exceeds the requested event queue length. USAGE
This operation is useful when the potential number of events that could be placed on the event queue changes dynamically. ATTRIBUTES
See attributes(5) for descriptions of the following attributes: +-----------------------------+-----------------------------+ | ATTRIBUTE TYPE | ATTRIBUTE VALUE | +-----------------------------+-----------------------------+ |Interface Stability |Standard: uDAPL, 1.1, 1.2 | +-----------------------------+-----------------------------+ |MT-Level |Unsafe | +-----------------------------+-----------------------------+ SEE ALSO
libdat(3LIB), attributes(5) SunOS 5.10 16 Jul 2004 dat_evd_resize(3DAT)
All times are GMT -4. The time now is 01:41 AM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy