Conference Papers

Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/28506

Browse

Subject: search.filters.subject.Software Defined Networking (SDN)

Search Results

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Minimizing Latency Due to Flow Table Overflow by Early Eviction of Flow Entries in SDN
    (Institute of Electrical and Electronics Engineers Inc., 2018) Khan, M.K.A.; Sah, V.K.; Mudgal, P.; Hegde, S.
    An Open vSwitch (OVS) stores flow entries in a flow table which is located on an extremely fast and efficient associative memory, Ternary Content Addressable Memory (TCAM). This TCAM is limited in size and therefore in scenarios of huge data traffic these may overflow and incur huge latency and packet loss. Our aim is to address these issues. In this paper we propose an early eviction strategy to evict the flow entries before the flow table overflows, based on analysis of couple of network parameters, namely number of hosts and packet arrival speed. We construct a relation between these parameters and eviction threshold, which is the value that decides the capacity of the flow table at which flow entries are to be evicted, to find the optimal threshold that minimizes the latency and packet loss. We have realized the results on star topology with single switch and multiple hosts, with a flow table limit of 10 entries. The simulation results show drastic decrease in latency and packet loss when flows are evicted early compared to when they are evicted after the table overflows. © 2018 IEEE.
  • No Thumbnail Available
    Item
    Next-Generation Technologies Empowered Future IoV
    (Institute of Electrical and Electronics Engineers Inc., 2022) Rajasekhar, K.; Kumar, R.; Manjappa, M.; Rammohana Reddy, G.
    The Internet of Vehicles is a network of connected automobiles that can communicate with one another. In the IoV environment, vehicles and sensors, pedestrians, mobile devices, and networking devices are the primary components; improving road safety, proper guidance while driving, autonomous driving, traffic management, infotainment, and entertainment. But with rapid growth in the number of automobiles, high vehicle mobility and networking devices; it isn't easy to achieve the requirements needed for the operation and management of IoV. Next-generation technologies such as 5th Generation (5G) and Software Defined Network (SDN) technologies can significantly address these challenges. These technologies will provide potential solutions for efficient management, enhance the performance of IoV. In this paper, we performed simulation with these technologies in the real world vehicular network along with performance evaluation. In comparison to traditional vehicle networks, our findings show that 5G and software-defined vehicular networks have a lower transmission delay and higher throughput. © 2022 IEEE.