H‐SOFT: a heuristic storage space optimisation algorithm for flow table of OpenFlow. (14th January 2014)

• Record Type: Journal Article
• Authors: Ge, Jingguo; Chen, Zhi; Wu, Yulei; E, Yuepeng; Limet, Sébastien; Smari, Waleed W.; Spalazzi, Luca; Hu, Jia; Gao, Jianliang
• Is Part Of: Concurrency and computation. Volume 27:Number 13(2015:Sep.)
• Page Start: 3497
• Page End: 3509
• Publication Date: 2014-01-14
• Subjects: Parallel processing (Electronic computers) -- Periodicals; Parallel computers -- Periodicals; 004.35
• DOI: 10.1002/cpe.3206
• Languages: English
• ISSNs: 1532-0626

Abstract

<abstract abstract-type="main"> <title>Summary</title> <p>OpenFlow has become the key standard and technology for software defined networking, which has been widely adopted in various environments. However, the global deployment of OpenFlow encountered several issues, such as the increasing number of fields and complex structure of flow entries, making the size of flow table in OpenFlow switches explosively grows, which results in hardware implementation difficulty. To this end, this paper presents the modelling on the minimisation for storage space of flow table and proposes a Heuristic Storage space Optimisation algorithm for Flow Table (H‐SOFT) to solve this optimisation problem. The H‐SOFT algorithm degrades the complex and high‐dimensional fields of a flow table into multiple flow tables with simple and low‐dimensional fields based on the coexistence and conflict relationships among fields to release the unused storage space due to blank fields. Extensive simulation experiments demonstrate that the H‐SOFT algorithm can effectively reduce the storage space of flow table. In particular, with frequent updates on flow entries, the storage space compression rate of flow table is stable and can achieve at ~70%. Moreover, in comparison with the optimal solution, the H‐SOFT algorithm can achieve the similar compression rate with much lower execution time. Copyright © 2014 John Wiley &amp; Sons, Ltd.</p> </abstract>