Resource provisioning system for mobile operator network's datacenters
Keywords:cloud computing, NFV, resource allocation, workload prediction, mobile network
AbstractThe problem of the mobile data traffic and the number of services growing becomes global, moreover, volume and frequency of control traffic transmitted through the network are increasing. Therefore there is a need for its effective management to ensure the quality of service required by users and optimal use of mobile network resources. In such circumstances the load on the server that is created in the process of the connection establishing and its serving has its considerations. Dynamic resource provisioning is a useful technique for handling the variations seen in communication systems workloads. Virtualization technology allows to implement this approach. An analytic model of a system would be attractive as it would be able to evaluate system characteristics under a wide range of conditions, and to be computed comparatively easily. It also can incorporate numerical optimization techniques for system design. In this paper the problem of provisioning system design for virtualized network functions is solved. This paper presents a novel approach to correctly allocate resources in data centers, such that SLA violations and energy consumption are minimized. Proposed approach first analyzes historical workload traces to identify long-term patterns that establish a "base" workload. Then it employs two techniques to dynamically allocate capacity: predictive and reactive provisioning. The combination of predictive and reactive provisioning achieves a significant improvement in meeting SLAs, conserving energy and reduces provisioning costs. The method for adapting the size of network function's resource allocation control interval is proposed. It provides dynamic configuration of the system, reducing excessive service data transmitted in the network and decreasing the load of network nodes. The workload service system model is built. It outlines a method of workload forecasting taking into account long-term accumulated statistics and recent trends observed in a network. It allows to get a rational level of management costs and final values of service quality. Experiments on the study of the proposed methods in the system of Mathcad are conducted.
Skulysh M. A. and Sulima S. V. (2015) Method of resource management in datacenters of mobile network operator, Radioelektronika i informatika, No 3, pp. 13-17.
Gandhi A., Chen Y., Gmach D., Arlitt M. and Marwah M. (2011) Minimizing Data Center SLA Violations and Power Consumption via Hybrid Resource Provisioning, Green Computing, pp. 1-8. DOI: 10.1109/IGCC.2011.6008611
Cai Z., Liu F., Xiao N., Liu Q. and Wang Z. (2010) Virtual network embedding for evolving networks, Global Telecommunications conference, pp. 1-5. DOI: 10.1109/GLOCOM.2010.5683160
Sun G., Yu H., Anand V. and Li L. (2013) A cost efficient framework and algorithm for embedding dynamic virtual network requests, Future Generation Computer Systems, Vol. 29, No. 5, pp. 1265-1277. DOI: 10.1016/j.future.2012.08.002
Chandra A., Gong W. and Shenoy P. (2003) Dynamic resource allocation for shared data centers using online measurements, 11th International Workshop on Quality of service, pp. 381-398. DOI: 10.1007/3-540-44884-5_21
Urgaonkar B., Shenoy P., Chandra A., Goyal P. and Wood T. (2008) Agile dynamic provisioning of multi-tier Internet applications, ACM Transactions on Autonomous and Adaptive Systems (TAAS), Vol. 3, No. 1, pp. 1-39. DOI: 10.1145/1342171.1342172
How to Cite
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).