Call for models (Model Checking Contest 2016)
MODEL CHECKING CONTEST 2016 - (1/2) - CALL FOR MODELS
GOALS
The Model Checking Contest (MCC) is a yearly event that assesses existing verification tools for concurrent systems on a set of models (i.e., benchmarks) proposed by the scientific community. All tools are compared on the same benchmarks and using the same computing platform, so that a fair comparison can be made, contrary to most scientific publications, in which different benchmarks are executed on different platforms.
The Model Checking Contest is organized in three steps:
- the present Call for Models (http://mcc.lip6.fr/cfm.php),
- the Call for Tools (http://mcc.lip6.fr/cft.php),
- and the Contest itself.
CALL FOR MODELS
At the core of the Model Checking Contest is a collection of models (http://mcc.lip6.fr/models.php) accumulated from the previous editions of the contest. This collection currently comprises 56 different models, which have been already used and cited in 30 scientific publications.
For the 2016 edition, we kindly ask the scientific community (beyond the developers of verification tools) to propose novel models. Each model should be representative of a non-trivial academic or industrial problem that involves concurrency aspects, and may belong to very diverse fields such as software or hardware design, networking, biology, etc.
All submitted models will be reviewed by the Model Board and we expect a dozen new models to be selected and added to the MCC collection. The authors of the selected models will be acknowledged on the Model Checking Contest web site.
All submitted models should be kept confidential until the list of selected models has been published. This is to ensure that the 2016 models are not known in advance by the tool developers participating in the Model Checking Contest.
By submitting a model, you explicitly allow the organizers of the Model
Checking Contest to freely use this model and publish it on the web.
Submitted models are expected to become part of the public domain. If your model is proprietary, do not submit it. Detailed information is available from http://mcc.lip6.fr/rules.php.
SUBMISSION DETAILS
A model can be either a "classical" P/T net, a Nested-Unit Petri net, or a colored Petri net (with/without guards on transitions, cartesian product on colors, and successor/predecessor functions). For a colored net, an equivalent "unfolded" P/T net may be provided as well.
A model may depend on one or many parameters that enable scaling (e.g., in the number of places, transitions, tokens, colors, etc.). To each parameterized model are associated as many "instances" (typically, between 2 and 20) as there are different combinations of values considered for the parameters of this model; each non-parameterized model has a single associated instance.
Detailed instructions for submission are given in the model submission kit available from http://mcc.lip6.fr/archives/ModelSubmissionKit.tar.gz.
To submit a model, four types of files should be provided:
- A PNML file describing the model. If the model is parameterized and exists in different instances, or if it is colored and has an equivalent P/T net, then several PNML files are provided. For information about the PNML format, please refer to the web site http://www.pnml.org and contact lom-messan.hillah@lip6.fr if help is needed. For information about Nested-Unit Petri nets, please refer to http://mcc.lip6.fr/nupn.php and contact hubert.garavel@inria.fr if help is needed.
- A LaTeX form that must be filled in to provide summary information about the model, its origin, its size, etc. See http://mcc.lip6.fr/models.php for examples of such a form.
- If possible, a picture of the model to be included in the LaTeX form.
- If possible, a set of relevant properties (typically, invariants, bounds, reachability, LTL or CTL formulas) that can be evaluated on this model. These properties can be expressed informally in English or given as XML files. Submitted properties, which are most useful to produce meaningful benchmarks, will be reviewed by the Formula Board.
IMPORTANT DATES
- Nov. 03, 2015: publication of the present Call for Models
- Jan. 30, 2016: deadline for model submission
- Apr. 15, 2016: individual notification of model acceptance/rejection
- June 1, 2016: on-line publication of the selected MCC'2016 models
- June 21, 2016: announcement of MCC'2016 results during the Petri Net conference (Torun, Poland)
COMMITTEES
General Chairs
- Didier Buchs - Univ. Geneva, Switzerland
- Fabrice Kordon - UPMC, France
Model Board
- Hubert Garavel - Inria/LIG, France
- Fabrice Kordon - UPMC, France
- Lom Messan Hillah - Univ. Paris Ouest, France
Formula Board
- Loig Jezequel - Univ. Nantes, France
- Emmanuel Paviot-Adet - Univ. Paris 5, France
- Cesar Rodriguez - Univ. Paris 13, France