Loading [a11y]/accessibility-menu.js
Concise Retrieval of Flow Statistics for Software-Defined Networks | IEEE Journals & Magazine | IEEE Xplore

Concise Retrieval of Flow Statistics for Software-Defined Networks


Abstract:

Software-defined networking (SDN) provides a new network architecture by decoupling control and data planes. By performing the tasks of the control plane in a centralized...Show More

Abstract:

Software-defined networking (SDN) provides a new network architecture by decoupling control and data planes. By performing the tasks of the control plane in a centralized manner, SDN enables the capability of programmability and flexibility to fulfill various types of network applications. The decoupling of control and data planes also brings additional messaging overhead and latency to degrade the efficiency of flow statistics collection, but timely and accurate flow statistics are critical for numerous SDN applications. In this article, we propose an effective scheme, flow statistics retrieval using cookie, to reduce the messaging overhead of acquiring traffic statistics from OpenFlow switches. Our scheme uses the cookie field of flow entries to designate a set of flows in a switch and avoid collecting flow statistics. Moreover, it provides the flexibility of assigning flow statistics retrieved from each switch to fulfill different requirements of messaging overhead. With the flexibility, we develop several algorithms of flow-statistics-to-switch assignment. The results of the performance evaluation show that our scheme can reduce both the messaging overhead and latency for retrieving flow statistics with only a few bits of the cookie field. The proposed algorithms can also balance the volume of flow statistics transmitted from a switch or minimize the transmission overhead.
Published in: IEEE Systems Journal ( Volume: 16, Issue: 1, March 2022)
Page(s): 554 - 565
Date of Publication: 05 April 2021

ISSN Information:

Funding Agency:


References

References is not available for this document.