Authors A.G. Feoktistov, R.O. Kostromin
Month, Year 11, 2016 @en
Index UDC 004.4’2+004.89
DOI 10.18522/2311-3103-2016-11-6575
Abstract Nowadays, the effective management of scalable applications for solving large fundamental and applied tasks in a heterogeneous distributed computing environment is a non-trivial problem. The promising approach to solving this problem is the use of multi-agent systems. Today, there is a wide range of frameworks for creating the multi-agent systems for various purposes. However, the known frameworks do not have all the necessary tools to automate the development of agents. They also do not provide a representation of knowledge about the subject domain of solved tasks and software/hardware infrastructure in the created multi-agent system. The aim of our study is to develop methods and tools to solve these problems. In the paper, we represent a comparative analysis of a number of the well-known multi-agent systems for the management of such applications. We also give a brief overview of frameworks for developing multi-agent systems and prove the selection of the system JADE as a basic tool. Our contribution is multifold. We formulate the principles of multi-agent systems for the management of scalable applications. The effective operation of the system is based on the integrated use of the computational, schematic and production knowledge as well as knowledge about the software/hardware infrastructure of the environment and administrative policies in its nodes. This knowledge is presented in the form of a conceptual model of the computing environment. We propose a methodology for creating of multi-agent systems. It is based on a bottom-up approach to the design of such systems. This approach uses the methods and tools for the synthesis of abstract programs which reflect the behavior of agents. These methods and tools are used to generate the code of agents. The developed tool for creation of subject-oriented multi-agent systems is the intelligent superstructure over tools of the system JADE. They significantly extend the capabilities of this system. Examples of solving problems of parametric synthesis of linear regulator for dynamic object and SAT-problems using scalable applications under the management of multi-agent system show the scalability and efficiency of distributed computing. This system is developed in accordance with the proposed principles of the organization of such systems.

Download PDF

Keywords Scalable application; distributed computing management; multi-agent system; development tools.
References 1. Qureshi M.B., Dehnavi M.M., Min-Allah N., Qureshi M.S., Hussain H., Rentifis I., Tziritas N., Loukopoulos T., Khan S.U., Xu C.Z., Zomaya A.Y. Survey on Grid Resource Allocation Mechanisms, Journal of Grid Computing, 2014, Vol. 12, No. 2, pp. 399-441.
2. Talia D. Cloud Computing and Software Agents: Towards Cloud Intelligent Services, Pro-ceedings of the 12th Workshop on Objects and Agent, 2011, pp. 2-6.
3. Leitao P., Inden U., Ruckemann C.-P. Parallelising Multi-agent Systems for High Performance Computing, Proceedings of the 3rd International Conference on Advanced Communications and Computation, 2013, pp. 1-6.
4. Kravari K., Bassiliades N. A Survey of Agent Platforms, Journal of Artificial Societies and Social Simulation, 2015, Vol. 18, No. 1, pp. 1-18.
5. Kumar A., Toussaint M., Zilberstein S. Scalable Multiagent Planning Using Probabilistic Inference, Proceedings of the 22nd International Joint Conference on Artificial Intelligence, 2011. pp. 2140-2146.
6. Amato A., Venticinque S. A Distributed Agent-Based Decision Support for Cloud Brokering, Scalable Computing: Practice and Experience, 2014, Vol. 15, No. 1, pp. 65-78.
7. Frey J., Tannenbaum T., Foster I., Livny M., Tuecke S. Condor-G: A Computation Manage-ment Agent for Multi-Institutional Grids, Journal of Cluster Computing, 2002, Vol. 5,
pp. 237-246.
8. YarKhan A., Dongarra J., Seymour K. GridSolve: The Evolution of a Network Enabled Solv-er, Grid-based problem solving environments, 2007, pp. 215-224.
9. Laxmi CH.V.T.E.V., Somasundaram K. Application Level Scheduling (AppLeS) in Grid with Quality of Service (QoS), International Journal of Grid Computing and Applications, 2014, Vol. 5, No. 2, pp. 1-10.
10. Shi Z. Advanced Artificial Intelligence, Hackensack: World scientific, 2011, 624 p.
11. Rezaee A., Rahmani A.M., Parsa S., Adabi S. A Multi-Agent Architecture for QoS Support in Grid Environment, Journal of Computer Science, 2008, Vol. 4, No. 3, pp. 225-231.
12. Singh A., Malhotra M. Agent Based Framework for Scalability in Cloud Computing, International Journal of Computer Science and Engineering Technology, 2012, Vol. 3, No. 4, pp. 41-45.
13. Bogdanova V.G., Bychkov I.V., Korsukov A.S., Oparin G.A., Feoktistov A.G. Multiagent Ap-proach to Controlling Distributed Computing in a Cluster Grid System, Journal of Computer and Systems Sciences International, 2014, Vol. 53, No. 5, pp. 713-722.
14. Bychkov I.V., Oparin G.A., Feoktistov A.G., Bogdanova V.G., Pashinin A.A. Service-oriented Multiagent Control of Distributed Computations, Automation and Remote Control, 2015,
Vol. 76, No. 11, pp. 2000-2010.
15. Bychkov I.V., Oparin G.A., Feoktistov A.G., Sidorov I.A., Bogdanova V.G., Gorsky S.A. Multiagent Control of Computational Systems on the Basis of Meta-Monitoring and Imitational Simulation, Optoelectronics, Instrumentation and Data Processing, 2016, Vol. 52,
No. 2, pp. 107-112.
16. Unland R., Klusch M., Calisti M. Software Agent-Based Applications, Platforms and Devel-opment Kits. Birkhauser Verlag, 2005, 455 p.
17. Oparin G.A., Feoktistov D.G. Planirovanie skhem resheniya zadach v instrumental'nom komplekse SATURN/PZ [Planning of Problem Solving Schemes in Framework SATURN/PZ], Komp'yuternaya logika, algebra i intellektnoe upravlenie: Trudy Vserossiyskoy shkoly [Computer logic, algebra and intelligence control: Proceedings of the National school]. Irkutsk: Izd-vo IrVTs SO RAN, 1994, Vol. 1, pp. 5-13.
18. Bellifemine F., Bergenti F., Caire G., Poggi A. Jade: a Java Agent Development Framework, Multiagent Systems, Artificial Societies, And Simulated Organizations: MultiAgent Pro-gramming, Eds. A. Bordini, M. Dastani, J. Dix and A. El Fallax Seghrouchni, 2006, Vol. 15, pp. 125-147.
19. Bychkov I.V., Oparin G.A., Feoktistov A.G., Bogdanova V.G., Sidorov I.A., Pashinin A.A. Mul'tiagentnyy podkhod k upravleniyu servis-orientirovannymi vysokoproizvoditel'nymi vychisleniyami [Multiagent approach to control service-oriented high performance computing], Vestnik komp'yuternykh i informatsionnykh tekhnologiy [Journal of Computer and Information Technology], 2016, No. 9, pp. 35-41.
20. Bychkov I.V., Oparin G.A., Bogdanova V.G., Gorskiy S.A., Pashinin A.A. Mul'tiagentnaya tekhnologiya avtomatizatsii parallel'nogo resheniya bulevykh uravneniy v raspredelennoy vychislitel'noy srede [A multiagent technology of automation for parallel solution of Boolean equations in a distributed computing environment], Vychislitel'nye tekhnologii [Computational technologies], 2016, Vol. 21, No. 3, pp. 5-17.

Comments are closed.