Monitoring system for load balancing of distributed computing system nodes based on smartphones

Sergey Balabaev, Sergey Lupin, Aye Min Thike

Abstract


In the contemporary era, researchers frequently encounter the necessity of executing high-performance computing without the availability of robust workstations or access to clusters. The processing power of a personal computer is insufficient for the resolution of resource-intensive applied tasks. However, this can be augmented by the integration of such devices as Android smartphones, which are equipped with multi-core processors and a substantial amount of random-access memory (RAM), enabling them to perform resource-intensive computations as a node of a distributed system. However, when heterogeneous devices are integrated into a single computing environment, even distribution of load between them results in low efficiency of the whole system. The solution to this problem is node load balancing, which takes into account the real, rather than peak, performance of the nodes of the distributed environment. This paper proposes a method for load balancing the nodes of a smartphone-based distributed computing system using the developed monitor to determine the real performance of smartphones. The device data is collated by the client application, assembled into a packet and transmitted to the server via the network or stored as a file within the smartphone's memory. The server then performs an analysis of the received message and presents the device characteristics in graphical form. The functionality of the monitor was evaluated through experimentation to ascertain the parameters of the smartphone when solving the test problem under varying temperature conditions, namely at room temperature, when blown by a fan, and in conditions of reduced temperature. The characteristics of the nodes were employed to achieve a state of equilibrium in the computing environment. The computational experiments demonstrated that the characteristics obtained with the assistance of the developed monitor enabled the elimination of the imbalance in the computing environment, an increase in its performance and a reduction in the computation time by a factor of 1.8. The developed monitor can be utilised to achieve equilibrium in the nodes of a distributed system comprising Android devices.

Full Text:

PDF (Russian)

References


Kurochkin I. et al. Using Mobile Devices in a Voluntary Distributed Computing Project to Solve Combinatorial Problems //Supercomputing: 7th Russian Supercomputing Days, RuSCDays 2021, Moscow, Russia, September 27–28, 2021, Revised Selected Papers 7. – Springer International Publishing, 2021. – S. 525-537.

Dolgov A.A. Razvorachivanie Grid-sistemy iz Mobil'nyh ustrojstv na platforme BOINC // Oblachnye i raspredelennye vychislitel'nye sistemy v jelektronnom upravlenii ORVSJeU-2022 v ramkah nacional'nogo suprekomp'juternogo foruma (NSKF-2022), 2022 s. 24-29

Phuc B. H. et al. Enhancing the performance of android applications on multi-core processors by selecting parallel configurations for source codes //2017 4th NAFOSTED Conference on Information and Computer Science. – IEEE, 2017. – S. 225-229.

Kumar T. U., Senthilkumar R. CWC*—Secured distributed computing using Android devices //2016 International Conference on Recent Trends in Information Technology (ICRTIT). – IEEE, 2016. – S. 1-7.

Mateeva G. et al. Some capabilities of android os for distributed computing //2021 Big Data, Knowledge and Control Systems Engineering (BdKCSE). – IEEE, 2021. – S. 1-6.

Balabaev S.A. Ocenka vychislitel'nyh vozmozhnostej mobil'nyh platform //28-ja Vserossijskaja mezhvuzovskaja nauchno-tehnicheskaja konferencija studentov i aspirantov «Mikrojelektronika i informatika - 2021», 2021.

Balabaev S.A., Lupin S.A. Ocenka vychislitel'nyh vozmozhnostej mobil'nyh ustrojstv na platforme OS Avrora //Mikrojelektronika i informatika - 2023. Materialy nauchno-tehnicheskoj konferencii, pp. S. 51-56. , 20 04 2023.

Balabaev S.A., Lupin S.A., Shakirov R.N. Vychislitel'nyj klaster na osnove smartfonov Android i mikrokomp'juterov Raspberry Pi //International Journal of Open Information Technologies. – 2022. – T. 10. – #. 7. – S. 86-93.

Acosta A., Almeida F. Parallel implementations of the particle filter algorithm for android mobile devices //2015 23rd Euromicro International Conference on Parallel, Distributed, and Network-Based Processing. – IEEE, 2015. – S. 244-247.

Acosta A., Almeida F. The particle filter algorithm: parallel implementations and performance analysis over Android mobile devices //Concurrency and Computation: Practice and Experience. – 2016. – T. 28. – #. 3. – S. 788-801.

Tang J. et al. PE-FedAvg: A Privacy-Enhanced Federated Learning for Distributed Android Malware Detection //2023 IEEE Intl Conf on Parallel & Distributed Processing with Applications, Big Data & Cloud Computing, Sustainable Computing & Communications, Social Computing & Networking (ISPA/BDCloud/SocialCom/SustainCom). – IEEE, 2023. – S. 474-481.

Salem H. Distributed computing system on a smartphones-based network //Software Technology: Methods and Tools: 51st International Conference, TOOLS 2019, Innopolis, Russia, October 15–17, 2019, Proceedings 51. – Springer International Publishing, 2019. – S. 313-325.

Hardy B., Phillips B. Android programming: the big nerd ranch guide. – Addison-Wesley Professional, 2013.

Vasil'ev A.N. Java. Ob"ektno-orientirovannoe programmirovanie. Uchebnoe posobie. Standart tret'ego pokolenija. – " Izdatel'skij dom"" Piter""", 2021

Aljohani M., Alam T. Design an M-learning framework for smart learning in ad hoc network of Android devices //2015 IEEE International Conference on Computational Intelligence and Computing Research (ICCIC). – IEEE, 2015. – S. 1-5.

Fain Y. Java programming 24-hour trainer. – John Wiley & Sons, 2011.


Refbacks

  • There are currently no refbacks.


Abava  Кибербезопасность IT Congress 2024

ISSN: 2307-8162