Representative Measurement Point Selection to Monitor Software-defined Networks

Hark, Rhaban; Ghanmi, Mohamed; Kar, Soummya; Richerzhagen, Nils; Rizk, Amr; Steinmetz, Ralf

doi:10.1109/lcn.2018.8638073

Cited by 4 publications

(11 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The above two sub‐sections evaluate the advancements of the node selection algorithm and the adaptive timeout mechanism respectively, and in this sub‐section, we present the results of comparing the proposed statistical collection scheme with the existing work in terms of communication overhead, flow completeness, and time consumption. The representative algorithms we choose for comparison include OpenTM, 9 CeMon, 8 FMS, 12 and RMPS 17 . OpenTM and CeMon are emblematic methods of performing per‐flow statistics and per‐switch statistics, respectively, and belong to pull‐based statistics methods.…”

Section: Discussionmentioning

confidence: 99%

“…Especially for per-flow statistics, query requests with the same number of flows will cause serious communication cost. Note that, compared with the push-based mechanism proposed in FMS 12 and RMPS, 17 our work is also greatly improved, because in the process of statistical collection, we not only avoid the redundancy caused by paying attention to all points in FMS 12 but also significantly reduce the overhead of repeated packet-in messages caused by fixed timeout values in RMPS. 17 We use the percentage of the number of complete flows in the statistical collection results to the total number of flows to express the accuracy of the algorithm.…”

Section: Performance Evaluation Of Overall Schemementioning

confidence: 97%

“…The representative algorithms we choose for comparison include OpenTM, 9 CeMon, 8 FMS, 12 and RMPS. 17 OpenTM and CeMon are emblematic methods of performing per-flow statistics and per-switch statistics, respectively, and belong to pull-based statistics methods. FMS 12 is selected as the representative of obtaining network status information in the original push-based statistics method, that is, using the default idle timeout value provided by OpenFlow.…”

Section: Performance Evaluation Of Overall Schemementioning

confidence: 99%

“…FMS 12 is selected as the representative of obtaining network status information in the original push-based statistics method, that is, using the default idle timeout value provided by OpenFlow. The statistical collection method proposed by RMPS 17 further covers the process of ranking the importance of nodes. We reproduced the main algorithms involved in the above references, verified the advantages of our proposed algorithm in terms of communication cost, and the simulation results are shown in Figure 11.…”

Section: Performance Evaluation Of Overall Schemementioning

confidence: 99%

“…For FMS, 12 we assume that the timeout value is set large enough to ensure that all flows trigger only one packet-in event. For RMPS, 17 we set its F I G U R E 9 Fixed idle_timeout mechanism vs. adaptive idle_timeout mechanism F I G U R 1 0 Average number of packet-in per-flow with different number of the flow initial idle_timeout value to be the same as the initial value in AFRT. Different from the existing works, 7 we assume in the experiment that all the algorithms have no knowledge of the flow information, which is more in line with the actual network state.…”

Section: Performance Evaluation Of Overall Schemementioning

confidence: 99%

See 4 more Smart Citations

Cost‐effective and accurate flow statistics collection in OpenFlow‐based SDN

et al. 2022

View full text Add to dashboard Cite

Summary Network resource scheduling and optimization require the acquisition of status information as a basis. High‐cost solutions lead to more resource consumption but only bring negligible benefits. To address this challenge, this paper proposes a novel statistics collection method adapted to OpenFlow‐based SDN, which can reduce the measurement cost while ensuring the statistical accuracy. First, based on the complex network theory, we propose multi‐path weighted closeness centrality (MWCC) to perform importance ranking on network switching nodes, which helps us select top‐k key nodes for statistical collection to reduce the overhead. Second, we propose an adaptive flow rule timeout mechanism AFRT. AFRT continuously optimizes the rule timeout values based on statistical results, further balancing flow table overhead and statistical accuracy. A series of simulation results on real network topologies verify the superiority of the proposed method in terms of communication cost, statistical accuracy, and time consumption, compared with the existing representative methods.

show abstract