2019/01/25 by Konstantinos Poularakis, Qiaofeng Qin, Poularakis, Konstantinos +9 · 1 citation
Computer Science · Engineering · #Advanced Memory and Neural Computing #Advanced Optical Network Technologies #FOS: Computer and information sciences #Networking and Internet Architecture (cs.NI) #Quantum-Dot Cellular Automata #Software-Defined Networks and 5G
paper · pdf · doi:10.48550/arxiv.1901.08936
openalex publication_date 2019/01/25 · openalex created_date 2022/07/30 · openalex updated_date 2026/07/28
Since the early development of Software-Defined Network (SDN) technology,\nresearchers have been concerned with the idea of physical distribution of the\ncontrol plane to address scalability and reliability challenges of centralized\ndesigns. However, having multiple controllers managing the network while\nmaintaining a "logically-centralized" network view brings additional\nchallenges. One such challenge is how to coordinate the management decisions\nmade by the controllers which is usually achieved by disseminating\nsynchronization messages in a peer-to-peer manner. While there exist many\narchitectures and protocols to ensure synchronized network views and drive\ncoordination among controllers, there is no systematic methodology for deciding\nthe optimal frequency (or rate) of message dissemination. In this paper, we\nfill this gap by introducing the SDN synchronization problem: how often to\nsynchronize the network views for each controller pair. We consider two\ndifferent objectives; first, the maximization of the number of controller pairs\nthat are synchronized, and second, the maximization of the performance of\napplications of interest which may be affected by the synchronization rate.\nUsing techniques from knapsack optimization and learning theory, we derive\nalgorithms with provable performance guarantees for each objective. Evaluation\nresults demonstrate significant benefits over baseline schemes that synchronize\nall controller pairs at equal rate.\n