WEB BASED STEERING OF PETRI NET MODELS

WEB BASED STEERING OF PETRI NET MODELS

Nada Elgewely¹ ;Mostafa Herajy² and El-Sayed Atlam3

1Department of Mathematics and Computer Science , Faculty of Science, Port Said University, Email: noody94@yahoo.com

2 Department of Mathematics and Computer Science, Faculty of Science, Port Said University, Email: mherajy@sci.psu.edu.eg

3Department of Mathematics, Faculty of Science, Tanta University, Email: satlam@yahoo.com

Key Words: Simulation of biochemical networks, Petri nets, Computational steering, Snoopy steering and simulation server, Web applications.

ABSTRACT

Petri nets play an important role on modeling biological as well as technical systems. They are useful in modeling and simulation of technical as well as biological models.Therefore, adopting an interactive technique like computational steering which allows users to change key parameters and variables during the simulation execution rather than waiting for the end of the simulation. However, the execution of Petri net models is usually carried out in a non-interactive way.

Computational steering provides more powers to the simulation of Petri nets. However, currently available tools to implement the steering of Petri nets models are not flexible to run differentplatforms and devices. Furthermore, the currently available simulators of Petri net models take a long time and the users are not able to change key parameters during the execution of the simulation.Petri nets and computational steering provide a powerful tool both on the representation and simulation level. Moreover, web applications facilitate the easy experiments with new models and technologies with minimal technical support. In this paper we introduce a framework to steer Petri net models via the web. This framework will present morefacilitiesto the usersin order to steer and simulate Petri net models without having to install anything locally on their computers. Additionally, this study proposes a flexible and versatile framework that supports the steering of Petri nets models on the web. The framework is based on Snoopy steering server and can run on a highly distributed manner.Moreover, it permits users to see the results of the model’s simulation immediately.

1. INTRODUCTION

Petri net playsan important role on modeling biological as well as technical systems[2][4][11]. However, the execution of Petri net models is usually performed in non-interactive way [10]. Thus, there is a need to remotely mange and analyze simulation output traces simultaneously while the simulation is in progress.

Computational steering provides more powers to simulate Petri net models. [5][6][7][8]. However, currently available tools work by sequentially simulate Petri nets models [10].This paper,introducesa framework that supports the steering of Petri net models on web. Moreover, it provides the ability to the users to simulate the models viaa web interface without the need for complicated installation packages on the user local machine.

   In addition, giving capability to the users to change on the values of key parameterswhen the simulations run by using computational steering technique and see the resultsimmediately. To facilitate the easy experiments to steers and simulates newPetri net models with minimal technical support. Subsequently, the overall time requiredto execute dry-lab experiments will decrease.

This paper,presents aframework for web-based steering, simulations andinteractively steer Petri nets as well as biochemical network models. There are many important features that are provided by using this web framework that facilitates the construction and executionof "wet-lab" experiments.

Web applications facilitate the easy experimentation with new models and technologies with minimal technical support. We are nowhaving the ability to steer Petri Net models via web rather than by an application. This means that we can try steering and simulating Petri net models without having to install anything locally on our computers.

Web based applications support kind of collaboration and distribution where system biologists need to interact with their experiments with each other by changing some key parameters. Such application based on approach that disconnects the users for their experiments during the simulation. Our aim is to overcome these limitations by integration computational steering with the simulation of biochemical reaction networks.

This paper is organized as follows: Section 2 introduces the Framework for simulation Petri net models discussing its interdependent individual components. Section 3 explains theimplementation for Snoopy steering framework by an application via web.Section 4 introduces collection data and experimental evaluation and considers the efficiency of developed framework. Finally, Section 5 sums up a conclusion and possible future work.

2. THE FRAMEWORK

The aim of this framework is to satisfy the user requirements, in order to improve the user experience while using Petri net simulation tools. This can be achievedvia the adaptation of the computational steering while simulating the Petri net (biological) models via the web [2][4].

2.1 Overview

Our application framework is implementedusing four components: graphical user interface (GUI), application programming interface (API) and web-based applications(WA)in addition to the steering server [4].To implement our framework, our approach has adopted Java applications as the main development language for the GUI construction.

Furthermore, this study implements the framework in order to make the users interact with their Petri nets simulation models via the web. We have to use the different programming technologies to implement the framework including java, php HTML, XML, C++ and other programming language [3].

The Web application should be able to communicate with S4 [Snoopy Steering and Simulation Server] therefore, we have to API library (Application Programming Interface) which is written using C++ and GUI (Graphical User Interface) in order to represent and visualize different Petri net models [5][9][10].

 Our framework is implemented by using different techniques. We use Java native interface programming framework. This interface programming has the ability to connect the simulator algorithms.

Figure 1: The Framework

Figure1shows the framework for Petri Nets (PN) and Computational Steering (CS) framework is provided and supported by steering server, steering programming interface (GUI, API) and the internal or the external simulators [11].

This framework will be available to run and simulate all the types and approaches for Petri nets models like the continuous, the stochastic and the hybrid Petri nets. The important features of the developed web framework are:

• Utilizes Petri nets as a modeling language to represent the Petri net models.

• Enabled the biologists and scientists to interactive with their simulations of their chemical networks and biological systems.

• Readable the representation of reaction networks.

• Distributed collaborative and interactive simulation of biochemical networks.

• Tight coupling of visualization and simulation for the Petri net models.

The framework is implemented for combining PN and CS where, the users can represent and interactive simulation of biochemical networks. CS needs PN and simulation of biochemical networks.The framework contains of four main components as follows:

2.2Steering Server

Steering Server (SS) is the central manager in our framework. It is the main component of this development framework. Server works as a communication traffic between the different components.The server enabled users to steer the running simulation. Moreover, Computational steering could share knowledge between users. The server is responsible for saving all information. For instance, the information which has saving by the server.

2.3Steering Graphical Programming Interface (GUI)

In order to give users, the ability to interact and control with the currently simulation, we have to use Steering GUI. Steering GUI provides with these facilities.

2.4Steering Application Programming Interface(API)

API is a library, the programmers used it to extend existing capabilities for adding new features or improving existing ones. We use API Library in our framework in order to make the design of the framework easy to implement. API could send and receive the simple data or compound types.

2.5Simulators

This framework includes two types of simulators internal simulators and external simulators. First, the internal simulators run as a part of the steering server. Subsequently, there is no need for additional work for using them directly. There are three categories of simulation types: Continuous Simulation, Stochastic Simulation and Hybrid Simulation. Each Type has a specific algorithm. We talk about the simulating Continuous Petri nets, Stochastic Petri Nets and Hybrid Petri nets in the previous section.

The second type of Simulators is the external simulators. This type of simulators is developed by the users. The users developed it in order to run a specific simulation algorithm or to reuse an existing code. Implementing like these codes by using our framework will save the user’s time.

3. IMPLEMENTATION

We adapt the distributed approach to design this framework to steer and simulate anybiological system or biochemical networks which are represented by Petri nets to besimulated via the web. Mathematical and computer science are required in order tomodel these systems.

Subsequently, Petri nets have been presented a dedicated methodand techniques to facilitate dry-lab experiments.This study has applied the developed framework and get a web application to steer and simulate models. Web based steering Petri nets application facilitate the easy experimentation with new models and technologies with minimal technical support.

Figure 2: Petri Net Web Application

We use the different programming technologies to implement the framework.

We adapt the distributed approach to design this framework to steer and simulate any biological system or biochemical networks which are represented by Petri nets to be simulated via the web.

 Mathematical and computer science are required in order to model these systems. Subsequently, Petri nets have been presented a dedicated methods and techniques to facilitate dry-lab experiments.

4. EXPERIMENTAL RESULTS

In this section our experimental results will achieve the following evaluation:

• Facilitates to steer and simulate Petri net models without having to install anything locally on their computers.
• Enables users to see the results of the model’s simulation immediately.
• The ability to change key parameters during the execution of the simulation.
• Saves more money biologists had been spent for their experiments.
• More capability in less time.
• Making different simulation for the same model.
• Executing different Petri net models with different simulators.
• Different views for Petri net models can be define to explore the simulation.
• Exploring the running Petri net models on the fly.
• Changing the values of the simulation parameters during the simulation is executed.
• Ability to control on the speed of the simulator.
• Connection to the simulation from any place or at any time.
• Sharing the simulation and the results with other users while model run.
• GUI-independent implementation.
•  Running the simulation on the web.

Finally, the framework is based on Snoopy steering server and can run on a highly distributed manner. Moreover, it permits users to see the results of the model’s simulation immediately.

5. CONCLUSION

This paper introduces a framework for combining computational steering and Petri nets in order to model and simulate biochemical networks via the web. As well as we presents a new web application for users and systems biologistsin order to allow them to simulate their Petri net models via the web.

As a result, for using web application, more facilitates will be available for the users.On the other hand, the users will have the ability to simulate Petri net models via the web instead of download or install any software programs locally on their computers.

Over and above,the users can run their dry lab experiments by using this web application as well as get the results immediately. In the other hand, the users have the ability to run their simulations for Petri net models as well visualize the results at the same time. Future work could focus in using Petri Nets in Arabic language. It is suggested that, for the future we will work for developing an application for steering Petri net models (steering GUI) on the mobile. We will continue in improving our framework in order to model and simulate Petri Net models via the mobile.

7. REFERENCES

[1] BranislavHru´z and MengChu Zhou(2007). Modeling and control of discrete-event dynamic systems: With Petri nets and other tools. Springer Science & Business Media.

[2] Jiacun Wang(2007). Petri nets for dynamic event-driven system modeling. Handbook of Dynamic System Modeling.

[3] Joe Saunders ; Caroline Lyon ; Frank Forster ; Chrystopher L Nehaniv and Kerstin Dautenhahn(2009). A constructivist approach to robot language learning via simulated babbling and holophrase extraction. In Artificial Life, 2009. ALife’09. IEEE Symposium on, pages 13–20. IEEE, 2009.

[4] Kees Van Hee ; Alexander Serebrenik ; Natalia Sidorova and Wil Van Der Aalst (2007). History-dependent petri nets. Petri Nets and Other Models of Concurrency–ICATPN ,, Pp: 164–183.

[5] MostafaHerajy and Monika Heiner Springer (2013). Towards a computational steering and petri nets framework for the modelling of biochemical reaction networks. In CS&P, Pp: 147–159.

[6] MostafaHerajy ; Martin Schwarick and Monika Heiner(2013). Hybrid Petri nets for modelling the eukaryotic cell cycle. In Transactions on Petri Nets and Other Models of Concurrency VIII, Pp: 123–141.

[7] MostafaHerajy and Monika HeinerSpringer (2014). A steering server for collaborative simulation of quantitative petri nets. In International Conference on Applications and Theory of Petri Nets and Concurrency, Pp: 374–384.

[8] Petri Kajonius and GunneGrankvist(2015).The impact of personality traits, values, and abilities on the view of uniqueness of consciousness. In Towards a Science of Consciousness, University of Helsinki, Finland, 9-13 June 2015, pages 315–no.

[9] Stephen E.Derenzo(2005).Practical interfacing in the laboratory. IEEE Aerospace and Electronic Systems Magazine, 20(9):34.

[10] Wil MP Van der Aalst(1998). The application of Petri nets to workflow management. Journal of circuits, systems, and computers, 8(1):21–66.

[11] WilfriedBrauer and Wolfgang Reisig(2009). Carl adam Petri and “Petri nets”. Fundamental Concepts in Computer Science, 3(5):129.

توجیه نماذج شبکات بیترى باستخدام الویب

ندى الجویلی¹ ، مصطفی حراجی²،السید عتلم³

1-     قسم الریاضیات وعلوم الحاسب کلیة العلوم جامعة بورسعید

2-     قسم الریاضیات وعلوم الحاسب کلیة العلوم جامعة بورسعید

3-     قسم الریاضیات کلیة العلوم جامعة طنطا

تعتبر شبکات بیترى من أدوات النمذجة  التى تحتوى على شکل رسومى بالإضافى إلى خلفیة تعتمد على قواعد الریاضیات. وقد نجحت شبکات بیترى منذ بدایتها فى نمذجة العدید من النظم التقنیة بالإضافة إلى استخدامها مؤخراً فى نمذجة ومحاکاة النظم البیولوجیة مما جعلها تتطور إلى العدید من الفئات الآخرى مثل شبکات بترى المستمرة والعشوائیة والمهجنة.

وبالرغم من الممیزات العدیدة التى تقدمها شبکات بیترى لنمذجة النظم بطریقة رسومیة، إلا أن عملیة المحاکاة مازالت تتم بشکل غیر تفاعلى. آى أن المستخدم لا یمکنه تغییر بعض المعاملات او المتغیرات أثناء تنفیذ المحاکاة. فى هذا السیاق تستخدم تقنیة التوجیه الحسابى للسماح للمستخدمین تغییر معاملات ومتغیرات المحاکاة أثناء تنفیذها بشکل تفاعلى.

ونهدف فى هذا البحث إلى تصمیم وتنفیذ إطار یتضمن کلاً من شبکات بیترى والتوجیه الحسابى لنمذجة النظم البیولوجیة بشکل تفاعلى من خلال الویب.