Spade software agents and their impact on hardware resources
Abstract
Eliminating the disadvantages of updating prices with a large number of products in the online store, we have found a solution in the application of the Smart Python Agent Development Environment (SPADE). The article presents the process of collecting data on SPADE and Openfire Server performance metrics in order to determine and analyze the consumption of system resources when connecting software agents with different types of behavior, in different numbers, as well as during interaction with a web application. In current study, JMeter is used as a tool for data collection and performance testing (including load and stress). Quantitative and qualitative methods of data analysis are used. When processing the collected values of indicators for the use of hardware resources, methods of mathematical statistics have been used to identify relations between indicators. To compare the behaviors of the SPADE software agent, to determine the effectiveness of one over the other, as well as to determine the effectiveness of using the agent interface in command line mode compared to its web counterpart in the form of a graphical user interface (in terms of performance), formulas for calculating the growth rate are used. During the study, the advantage of SPADE in the speed of program code execution; the difference in performance between agent behaviors, as well as between agent web user interface (AWUI) and command line interface (CLI) modes; features of using the CLI mode of the agent for interactive user interaction with the application in order to quickly fix errors that occur during the interaction of the agent with the web application have been determined. Integration of SPADE agents into the pricing process has practical implications for retailers, opening up opportunities to study and develop new tools for subsequent application in solving specific problems
Keywords
XMPP; Python; behavior; web; JMeter; CPU; RAM
References
[1] Abbas, R., Sultan, Z., & Bhatti, S.N. (2017). Comparative study of load testing tools: Apache JMeter, HP LoadRunner, Microsoft Visual Studio (TFS), Siege. In International Conference on Communication Technologies (ComTech) (pp. 39-44). Rawalpindi, Pakistan. doi: 10.1109/COMTECH.2017.8065747.
[2] about:memory – Firefox Source Docs documentation. Retrieved from https://firefox-sourcedocs.mozilla.org/performance/memory/about_colon_memory.html.
[3] Advanced Behaviours – SPADE 3.3.0 documentation. Retrieved from https://spademas.readthedocs.io/en/latest/behaviours.html.
[4] Ajitha, S., Mithun, G., & Kumar, T.V.S. (2016). Optimal travel management using software agent. In International Conference on Circuits, Controls, Communications and Computing (I4C). Bangalore, India, 1-4. doi: 10.1109/CIMCA.2016.8053289.
[5] Ali, S., & Chernenko, A. (2020). Performance Testing – Microsoft Dynamics 365 Finance and Operations For sales order creation web service by deploying Blazemeter and JMeter, LAP LAMBERT Academic Publishing, July 2020, 56 p. ISBN: 978-620-2-67341-9.
[6] Apache JMeter. Retrieved from https://jmeter.apache.org.
[7] Dilshan de Silva et al. (2023). Evaluating the effectiveness of different software testing frameworks on software quality. PREPRINT (V. 1). doi: 10.21203/rs.3.rs-2928368/v1. Documentation: JMeter-Plugins.org. Retrieved from https://jmeter-plugins.org/wiki/ PerfMon/.
[8] Documentation: JMeter-Plugins.org. Retrieved from https://jmeter-plugins.org/wiki/ DummySampler/.
[9] Donancio, H., Casals, A., & Brandão, A.A. (2019). Exposing agents as web services: A case study using JADE and SPADE. Retrieved from https://gsigma.ufsc.br/wesaac2019/paper/ WESAAC_2019_paper_22.pdf.
[10] Dueñas, J.A. et al. (2021). Magnetic influence on water evaporation rate: An empirical triadic model. Journal of Magnetism and Magnetic Materials, 539, 168377. ISSN 0304-8853. doi: 10.1016/j.jmmm.2021.168377.
[11] GitHub - hyperic/sigar: System Information Gatherer and Reporter. Retrieved from https://github.com/hyperic/sigar.
[12] GitHub - javipalanca/spade_bdi: Plugin for SPADE 3 MAS platform to implement BDI Agents. Retrieved from https://github.com/javipalanca/spade_bdi.
[13] GitHub - undera/perfmon-agent: Server metrics fetching agent, based on SIGAR. Retrieved from https://github.com/undera/perfmon-agent.
[14] Graffius, S.M. (2023). Leverage the Power of the Minimum Viable Product (MVP), Retrieved from https://doi.org/10.13140/RG.2.2.24064.20486.
[15] Holgado-Terriza, J.A., Pico-Valencia, P., & Garach-Hinojosa, A. (2020). A gateway for enabling uniform communication among inter-platform JADE agents, IOS Press, Intelligent Environments, 28, 82-91. doi: 10.3233/AISE200027.
[16] Jubilson, A.E. et al. (2016). Revolution in e-commerce by the usage of software agents. International Journal of Advanced Computing and Electronics Technology (IJACET), 3(5). Retrieved from https://troindia.in/journal/ijacet/vol3iss5/16-19.pdf.
[17] Lyu, G., Fazlirad, A., & Brennan, R.W. (2020). Multi-agent modeling of cyber-physical systems for IEC 61499 based distributed automation. Procedia Manufacturing, 51, 12001206, ISSN 2351-9789. doi: 10.1016/j.promfg.2020.10.168.
[18] Mahadevan, R. et al. (2022). Payday loans – blessing or growth suppressor? Machine learning analysis. General Economics (econ.GN), Machine Learning (cs.LG). doi: 10.48550/arXiv.2205.15320.
[19] Matam, S., & Jain, J. (2017). JMeter plugins. In Pro Apache JMeter (pp. 211-219). Apress, Berkeley, CA. doi: 10.1007/978-1-4842-2961-3_9.
[20] Mokhamd, H., Arief, G., & Yoan, I. (2023). Analysis of application performance testing using load testing and stress testing methods in API service. Journal of Sisfotek Global, 13(1), 28-34. doi: 10.38101/sisfotek.v13i1.2656.
[21] Nordeen, A. (2020). Learn Software Testing in 24 Hours: Definitive Guide to Learn Software Testing for Beginners, N.p., Guru99, 291 p.
[22] Pal, C.-V. et al. (2020). A review of platforms for the development of agent systems. Multiagent Systems (cs.MA), 40 p. doi: 10.48550/arXiv.2007.08961.
[23] Palanca, J. (2017). SPADE: Agents based on XMPP. Retrieved from https://www.slideshare.net/JavierPalanca/spade-agents-based-on-xmpp-82102493.
[24] Palanca, J. (2018). SPADE Documentation – spade.mas.pdf, v3.0.0, Aug. 17. Retrieved from https://buildmedia.readthedocs.org/media/pdf/spade-mas/feature-3.0/spade-mas.pdf.
[25] Palanca, J. (2023). Spade Documentation, v. 3.3.0, Jun. 13. Retrieved from https://spademas.readthedocs.io/_/downloads/en/latest/pdf/.
[26] Palanca, J. et al. (2022). A flexible agent architecture in SPADE. In Advances in Practical Applications of Agents, Multi-Agent Systems, and Complex Systems Simulation. The PAAMS Collection: 20th International Conference (pp. 320-331), 13616. doi: 10.1007/978-3-03118192-4_26.
[27] Palanca, J. et al. (2023). Flexible agent architecture: Mixing reactive and deliberative behaviors in SPADE. Electronics, 12(3), 659. doi: 10.3390/electronics12030659.
[28] Palanca, J., Terrasa, A., Julian, V., & Carrascosa, C. (2020). SPADE 3: Supporting the new generation of multi-agent systems. IEEE Access, 8, 182537-182549. doi: 10.1109/ACCESS.2020.3027357.
[29] Pérez, S.F. (2023). Spade-BDI Documentation. Release 0.3.0, Jun 13. Retrieved from https://spade-bdi.readthedocs.io/_/downloads/en/latest/pdf/.
[30] Pohilko, A., JMeter Plugins - More Powerful Load Testing with JMeter Plugins. Retrieved from https://www.methodsandtools.com/tools/jmeterplugins.php.
[31] Python. (n.d.).Retrieved from https://www.python.org.
[32] Slhoub, Kh.A.M. (2018). Standardizing the Requirements Specification of Multi-Agent Systems. Florida Institute of Technology, 143. Retrieved from https://repository.lib.fit.edu/ bitstream/handle/11141/2608/SLHOUB-DISSERTATION-2018.pdf.
[33] SPADE – SPADE 3.3.0 documentation. (n.d.). Retrieved from https://spade-mas.readthedocs.io/ en/develop/readme.html.
[34] SPADE. (n.d.). Retrieved from https://pypi.org/project/spade/.
[35] Spade-BDI – Spade-BDI 0.3.0 documentation. (n.d.). Retrieved from https://spadebdi.readthedocs.io/en/latest/readme.html.
[36] The SPADE Agent Model – SPADE 3.3.0 Documentation. Retrieved from https://spademas.readthedocs.io/en/latest/model.html.
[37] Umbreen, J., Mirza, M.Z., Ahmad, Y., & Naseem, A. (2022). Assessing the role of minimum viable products in digital startups. In IEEE International Conference on Industrial Engineering and Engineering Management (IEEM) (pp. 1073-1077). Kuala Lumpur, Malaysia. doi: 10.1109/IEEM55944.2022.9989653.
[38] XMPP Load Testing - The Ultimate Guide. Retrieved from https://www.blazemeter.com/ blog/xmpp-testing.
[39] Zelenko, E. (2022). Determining the correlation between datasets for calculation of the retail price when using software agents. Management of Development of Complex Systems, 50, 102-105. doi: 10.32347/2412-9933.2022.50.102-105.
[40] Zelenko, E., & Kataieva, Y. (2023). Overview of methods and software for pricing. In Sworld-Us Conference Proceedings (pp. 23-27), 1(usc17-01). doi: 10.30888/27092267.2023-17-01-023.
[41] Zelenko, E., & Kataieva, Ye.Yu. (2023). Classification and synthesis of the main dropshipping disadvantages to eliminate them using software agents. Electronic Modeling, 45(2), 115-122. doi: 10.15407/emodel.45.02.115.