To analysis of a two-buffer queuing system with cross-type service and additional penalties

Cover Page

Abstract


The concept of cloud computing was created to better preserve user privacy and data storage security. However, the resources allocated for processing this data must be optimally allocated. The problem of optimal resource management in the loud computing environment is described in many scientific publications. To solve the problems of optimality of the distribution of resources of systems, you can use the construction and analysis of QS. We conduct an analysis of two-buffer queuing system with cross-type service and additional penalties, based on the literature reviewed in the article. This allows us to assess how suitable the model presented in the article is for application to cloud computing. For a given system different options for selecting applications from queues are possible, queue numbers, therefore, the intensities of transitions between the states of the system will change. For this, the system has a choice policy that allows the system to decide how to behave depending on its state. There are four components of such selection management models, which is a stationary policy for selecting a queue number to service a ticket on a vacated virtual machine each time immediately before service ends. A simulation model was built for numerical analysis. The results obtained indicate that requests are practically not delayed in the queue of the presented QS, and therefore the policy for a given model can be considered optimal. Although Poisson flow is the simplest for simulation, it is quite acceptable for performance evaluation. In the future, it is planned to conduct several more experiments for different values of the intensity of requests and various types of incoming flows.


Full Text

1. Introduction In order to preserve and protect the users confidential data of computing resources, the concept of Cloud Computing was developed as a way to provide secure storage and processing of data for companies and individuals. Cloud Computing includes not only programs and applications delivered as services over the Internet, but also the hardware and system software in the data centers that provide those services. This technology has five main charac- teristics [1]: on-demand self-service, broad network access, resource pooling, rapid elasticity and measured service. In addition, Cloud Computing includes three main types of services: Infrastructure as a Service, Platform as a Ser- vices, and Software as a Service [2]. There are four different ways to use this technology: Public Cloud, Private Cloud, Community Cloud, and Hybrid Cloud. Nowadays the Cloud Computing model has taken on an increasingly promi- nent role in a variety of IT-environments, where service providers seek to meet the needs of their customers and improve their competitive position. The in- crease in the number of users and the expansion of the services provided has led to the need for more storage space. As a result, service providers must work to increase the bandwidth of online data centers. Cloud Computing has become an integral part of maintaining high performance to improve com- petitiveness [3]. It is the fastest growing technology, and therefore, there are some challenges for developers and for those who use them. Let’s consider some tasks: § tasks of distribution and use of resources; § model of calculations MapReduce (model of parallel computing over very large amounts of data) [4]; § protection of cloud infrastructure [5]; § ensuring the reliability of the work of many servers [6]; § homomorphic codes (a form of encryption); § identification of spam pages [7]; § organization of information search. The problem of optimal resource management in the Сloud Сomputing environment is described in many scientific publications. As known, one of the approaches to solving this problem is the construction of Queuing Systems (QS). To analyze the distribution of resources and develop an optimal method of performance management, in [8] a multiservice QS of a Cloud Computing model with the same type of tasks and identical servers is investigated. The optimization criterion is the minimization of the ratio of the average queue length to the number of lost tasks. It should be noted that the efficient operation of such a network presupposes the ability to flexibly respond to changes in the demand for computing power by turning on/off machines. Therefore, for a heterogeneous environment of Cloud Computing virtual machines the open Jackson queue network model was proposed in [9], which allows solving the problem of scaling the number of virtual machines. To solve this problem the architecture of an elastic system of dynamic resource management with several queues is presented in [10]. The model of an open system with message queues is presented in [11], where reliability is guaranteed due to the mechanism for optimizing the timeout duration, which 160 DCM&ACS. 2021, 29 (2) 158-172 does not allow the loss of a single message. The cloud architecture on e- health platforms in medical centers was studied in [12], where a model of two sequential

About the authors

Irina A. Kochetkova

Peoples’ Friendship University of Russia (RUDN University); Institute of Informatics Problems, FRC CSC RAS

Author for correspondence.
Email: gudkova-ia@rudn.ru
ORCID iD: 0000-0002-1594-427X
6, Miklukho-Maklaya St., Moscow, 117198, Russian Federation; 44-2, Vavilova St., Moscow, 119333, Russian Federation

Candidate of Physical and Mathematical Sciences, assistant professor of Department of Applied Probabil- ity and Informatics of Peoples’ Friendship University of Russia (RUDN University); Senior Researcher of Institute of Informatics Problems of Federal Research Center “Computer Science and Control” Russian Academy of Sciences

Anastasia S. Vlaskina

Peoples’ Friendship University of Russia (RUDN University)

Email: vlaskina.anastasia@yandex.ru
ORCID iD: 0000-0001-6453-814X
6, Miklukho-Maklaya St., Moscow, 117198, Russian Federation

PHD student of Department of Applied Probability and Informatics

Dmitriy V. Efrosinin

Peoples’ Friendship University of Russia (RUDN University); Johannes Kepler Universität Linz

Email: dmitry.efrosinin@jku.at
ORCID iD: 0000-0002-0902-6640
6, Miklukho-Maklaya St., Moscow, 117198, Russian Federation; 69, Altenberger Straße, Linz, 4040, Austria

Doctor of Science in physics and mathematics, associate professor of Johannes Kepler Universitaet Linz; associate professor of Peoples’ Friendship University of Russia (RUDN University)

Abdukodir A. Khakimov

Peoples’ Friendship University of Russia (RUDN University)

Email: khakimov-aa@rudn.ru
ORCID iD: 0000-0003-2362-3270
6, Miklukho-Maklaya St., Moscow, 117198, Russian Federation

Junior researcher of Department of Applied Probability and Informatics

Sofiya A. Burtseva

Peoples’ Friendship University of Russia (RUDN University)

Email: sofiya_burceva@inbox.ru
ORCID iD: 0000-0003-4305-7050
6, Miklukho-Maklaya St., Moscow, 117198, Russian Federation

Master student of Department of Applied Probability and Informatics

References

  1. M. Armbrust, A. Fox, R. Griffith, A. D. Joseph, R. Katz, A. Konwinski, and M. Zaharia, “A view of cloud computing,” Communications of the ACM, vol. 4, no. 53, pp. 50-58, 2010. doi: 10.1145/1721654.1721672.
  2. N. Taleb and E. A. Mohamed, “Cloud computing trends: a literature review,” Academic Journal of Interdisciplinary Studies, vol. 1, no. 9, pp. 91-104, 2020. doi: 10.36941/ajis-2020-0008.
  3. I. Baldini, P. Castro, K. Chang, P. Cheng, S. Fink, V. Ishakian, and P. Suter, “Serverless computing: current trends and open problems,” Research Advances in Cloud Computing, pp. 1-20, 2017. doi: 10.1007/978-981-10-5026-8_1.
  4. V. Sontakke and R. B. Dayanand, “Optimization of Hadoop MapReduce Model in cloud Computing Environment,” Proceedings of the 2nd International Conference on Smart Systems and Inventive Technology, ICSSIT 8987823, 2019, pp. 510-515. doi: 10.1109/ICSSIT46314.2019.8987823.
  5. A. Yashwanth Reddy and R. P. Singh, “Design and development of multi tenancy in cloud: security issues,” International Journal of Scientific and Technology Research, vol. 3, no. 9, pp. 694-697, 2020.
  6. B. Kalyani and K. Rao, “Assessment of physical server reliability in multi cloud computing system,” AIP Conference Proceedings 1952,020045, 2018. doi: 10.1063/1.5032007.
  7. Y. Li, Y. Xu, and J. Chen, “Research on spam pages identification in search service based on cloud computing,” Huazhong Keji Daxue Xuebao (Ziran Kexue Ban)/Journal of Huazhong University of Science and Technology (Natural Science Edition), vol. 1, no. 40, pp. 249-253, 2012.
  8. A. Madankan, A. Delavarkhalafi, S. M. Karbassi, and F. Adibnia, “Resource allocation in cloud computing via optimal control to queuing system,” Bulletin of the South Ural State University, Series: Mathematical Modelling, Programming and Computer Software, vol. 4, no. 12, pp. 67-81, 2019. doi: 10.14529/mmp190405.
  9. C. N. Khac, K. B. Thanh, H. H. Dac, S. N. Hong, V. P. Tran, and H. T. Cong, “An Open Jackson Network Model for Heterogeneous Infrastructure as a Service on Cloud Computing,” International Journal of Computer Networks and Communications, vol. 1, no. 11, pp. 63-80, 2019. doi: 10.5121/ijcnc.2019.11104.
  10. Z. Cheng, H. Li, Q. Huang, Y. Cheng, and G. Chen, “Research on elastic resource management for multi-queue under cloud computing environment,” Journal of Physics: Conference Series, vol. 9, no. 898, 2017. doi: 10.1088/1742-6596/898/9/092003.
  11. L. Jing, C. Yidong, and M. Yan, “Modeling Message Queueing Services with Reliability Guarantee in Cloud Computing Environment Using Colored Petri Net,” Mathematical Problems in Engineering, 2015. doi: 10.1155/2015/383846.
  12. S. Kannan and S. Ramakrishnan, “Performance Analysis of Cloud Computing in Healthcare System Using Tandem Queues,” International Journal of Intelligent Engineering and Systems, vol. 4, no. 10, pp. 256-264, 2017. doi: 10.22266/ijies2017.0831.27.
  13. D. Efrosinin, I. Gudkova, and N. Stepanova, “Algorithmic analysis of a two-class multi-server heterogeneous queuing system with a controllable cross-connectivity,” Lecture Notes in Computer Science, vol. 12023, 2020. doi: 10.1007/978-3-030-62885-7_1.

Statistics

Views

Abstract - 99

PDF (English) - 33

Cited-By


PlumX

Dimensions


Copyright (c) 2021 Kochetkova I.A., Vlaskina A.S., Efrosinin D.V., Khakimov A.A., Burtseva S.A.

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

This website uses cookies

You consent to our cookies if you continue to use our website.

About Cookies