A new smart networking architecture for container network functions

Atici G., Boluk P.

TURKISH JOURNAL OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCES, vol.28, no.6, pp.3368-3388, 2020 (SCI-Expanded) identifier identifier identifier

  • Publication Type: Article / Article
  • Volume: 28 Issue: 6
  • Publication Date: 2020
  • Doi Number: 10.3906/elk-2002-68
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Applied Science & Technology Source, Compendex, Computer & Applied Sciences, INSPEC, TR DİZİN (ULAKBİM)
  • Page Numbers: pp.3368-3388
  • Istanbul University Affiliated: No


5G slices have challenging application demands from a wide variety of fields including high bandwidth, low latency and reliability. The requirements of the container network functions which are used in telecommunications are different from any other cloud native IT applications as they are used for data plane packet processing functions, together with control, signalling and media processing which have critical processing requirements. This study aims to discover high performing container networking solution by considering traffic loads and application types. The behaviour of several container cluster networking solutions - Flannel, Weave, Libnetwork, Open Virtual Networking for Open vSwitch and Calico - are explored with regard to the most commonly used container network functions in the form of MongoDB and web access. Evaluations show that none of the solutions provide a high throughput for all types of workload under optimum or heavy load situations. Hence, this research presents the view that traditional container networking implementation methods may not fulfill the container network functions' networking performance requirements. This is because container networking performance changes dynamically, depending on traffic load and application types. To overcome this problem, a new smart container networking architecture is proposed which allows containers to use several container networking solutions dynamically in conjunction with container monitoring tools. Eventually, the primary implementation of proposed architecture has been performed and evaluated. This research shows that proposed smart architecture delivers promising results compared to traditional implementation methods, in case appropriate decision is made during dynamic interface selection process.