Flow Management Mechanism in Software-Defined Network
Source abstract
Abstract
Software-defined networking (SDN) is a paradigm shift in modern networking. However, centralised controller architecture in SDN imposed flow setup overhead issue as the control plane handles all flows regardless of size and priority. Existing frameworks strictly reduce control plane overhead and it does not focus on rule placement of the flows itself. Furthermore, existing frameworks do not focus on managing elephant flows like RTSP. Thus, the proposed mechanism will use the flow statistics gathering method such as random packet sampling to determine elephant flow and microflow via a pre-defined threshold. This mechanism will ensure that the control plane works at an optimum workload because the controller only manages elephant flows via reactive routing and rule placement respectively. Reactive routing has reduced link bandwidth usage below the pre-defined threshold. Furthermore, rule placement has increased average throughput and total transfer to 238%. Meanwhile, the data plane switches will be able to forward microflows via multipath wildcard routing without invoking controller in greater responding time by 85 ms faster in two Transmission Control Protocol (TCP) traffic and achieved 11% and 12% higher total transfer size and throughput respectively. Hence, the controller’s workload reduced significantly to 48% in two TCP traffic.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Software-Defined Networks and 5G · Network Traffic and Congestion Control · Interconnection Networks and Systems
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (20)
Show source-provided references
1. Real-time QoS-aware routing scheme in SDN-based robotic cyber-physical systems
2. Optimal 5G network slicing using machine learning and deep learning concepts10.1016/j.csi.2021.103518 ↗
3. Pilot decontamination using asynchronous fractional pilot scheduling in massive MIMO systems10.3390/s20216213 ↗
4. DevoFlow: Scaling flow management for high-performance networks
5. An optimized flow management mechanism in OpenFlow network
6. Network-wide traffic visibility in OF@TEIN SDN testbed using sFlow
7. sFlow-RT@ ONLINE, [Online]. Available: http://www.inmon.com/products/sFlow-RT.php.
8. Improved bounds on nonblocking 3-stage clos network10.1137/060656413 ↗
9. Optimal bandwidth reservation in hose-model VPNs with multi-path routing
10. Maximum throughput routing of traffic in the hose model
11. Scalable flow-based networking with DIFANE10.1145/1851275.1851224 ↗
12. Kandoo: A framework for efficient and scalable offloading of control applications
13. ASIC: An architecture for scalable intra-domain control in OpenFlow
14. Hyperflow: A distributed control plane for openflow
15. Koponen · OSDI · 2010 · 10 · 1
16. Using memcached for data distribution in industrial environment
17. Stribling · NSDI · 2009 · 9 · 43
18. Distributed controller clustering in software defined networks10.1371/journal.pone.0174715 ↗
19. Rules placement problem in OpenFlow networks: A survey10.1109/COMST.2015.2506984 ↗
20. Optimizing rules placement in openflow networks: Trading routing for better efficiency