[gme-users] Some interesting results in model scaling and GME limitations

[Jane Lin]

MessageHi,

We performed some model scaling experiments recently, which might be of
interest to some of you. These point to potential limitations of GME with
respect to the size of models that can be scaled at once in batch. We were
using the latest publicly available version of GME for this.

There is a Petri Net modeling language that Dr. Andy Gokhale created (called
the SRNML). In this language, an "event type" for the Reactor pattern is
modeled as a series of places and transitions. We wanted to use our model
transformation engine (called C-SAW) to push the GME to its limits in terms
of the number of modeling elements that could be added.

We created model transformations to scale from 2 event types to 438 event
types. This represents about 4,400 modeling elements (basically, all of
these were atoms and connections) added in batch all at once. On the
addition of the 439 event type, the GME crashes.

We confirmed this situation on several different machines, so it seems to be
a characteristic of GME that is independent of the platform.

An interesting observation was made: it was possible to add 500 event types
if the replication was performed as two steps of 250 event types (rather
than 500 at once, which would crash at 438). This seems to suggest that a
GME interpreter may add modeling elements at a speed that eventually GME
could not process. It would be good to get feedback on this from GME
developers on this observation.

Part of this was done for a journal paper we just submitted on model
transformations as replicators. In case anyone is interested in looking at
this work, an URL for the submitted paper (currently under review by SoSym)
is at:

http://www.cis.uab.edu/gray/Pubs/sosym-models-2006.pdf

Thanks

Jane, Jeff, and Jing at UAB

---------------------------------------------------------------------------------
Yuehua (Jane) Lin
Department of Computer and Information Sciences
University of Alabama at Birmingham
liny at cis.uab.edu 
http://www.cis.uab.edu/liny

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://list.isis.vanderbilt.edu/pipermail/gme-users/attachments/20060220/47f53a45/attachment.htm