File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1016/j.comnet.2019.107007
- Scopus: eid_2-s2.0-85075263872
- WOS: WOS:000510524600005
- Find via
Supplementary
- Citations:
- Appears in Collections:
Article: Flow monitoring scheme design in SDN
Title | Flow monitoring scheme design in SDN |
---|---|
Authors | |
Keywords | Critical Column First Flow monitoring scheme Lowest-cost Switch first Match structure SDN |
Issue Date | 11-Feb-2020 |
Publisher | Elsevier |
Citation | Computer Networks, 2020, v. 167 How to Cite? |
Abstract | An SDN controller can collect all flow statistics by periodically polling switches in the network. Each polling request contains a match structure identifying one or many flow entries at a switch. A flow is covered if its statistics is collected. In this paper, a new polling mechanism called poll-some is introduced. As compared to the two existing mechanisms, poll-single and poll-all, poll-some aims at collecting the statistics of multiple not-yet-covered flows at a switch using a minimum number of match structures. The problem of finding the minimum number of match structures to cover all the not-yet-covered flows at a switch, or Minimum Match Structure (MMS) problem, is then formulated. An efficient algorithm called Critical Column First (CCF) is designed to solve it. Based on CCF, a new flow monitoring scheme called Lowest-cost Switch First (LSF) is proposed to minimize the controller-switch communication bandwidth cost incurred in collecting flow statistics. LSF consists of two phases, finding the switch to be polled next and determining the best polling mechanism. As compared with an existing greedy algorithm, simulation results show that our LSF not only runs faster but also cut down the bandwidth cost by about 30%. |
Persistent Identifier | http://hdl.handle.net/10722/339776 |
ISSN | 2021 Impact Factor: 5.493 2020 SCImago Journal Rankings: 0.798 |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Yang, Z | - |
dc.contributor.author | Yeung, Kwan L | - |
dc.date.accessioned | 2024-03-11T10:39:14Z | - |
dc.date.available | 2024-03-11T10:39:14Z | - |
dc.date.issued | 2020-02-11 | - |
dc.identifier.citation | Computer Networks, 2020, v. 167 | - |
dc.identifier.issn | 1389-1286 | - |
dc.identifier.uri | http://hdl.handle.net/10722/339776 | - |
dc.description.abstract | <p>An <a href="https://www.sciencedirect.com/topics/engineering/software-defined-networking" title="Learn more about SDN from ScienceDirect's AI-generated Topic Pages">SDN</a> <a href="https://www.sciencedirect.com/topics/computer-science/can-controller" title="Learn more about controller can from ScienceDirect's AI-generated Topic Pages">controller can</a> collect all flow statistics by periodically polling switches in the network. Each polling request contains a match structure identifying one or many flow entries at a switch. A flow is covered if its statistics is collected. In this paper, a new polling mechanism called <em>poll-some</em> is introduced. As compared to the two existing mechanisms, <em>poll-single</em> and <em>poll-all, poll-some</em> aims at collecting the statistics of multiple not-yet-covered flows at a switch using a minimum number of match structures. The problem of finding the minimum number of match structures to cover all the not-yet-covered flows at a switch, or Minimum Match Structure (MMS) problem, is then formulated. An efficient algorithm called Critical Column First (CCF) is designed to solve it. Based on CCF, a new flow monitoring scheme called Lowest-cost Switch First (LSF) is proposed to minimize the controller-switch <a href="https://www.sciencedirect.com/topics/computer-science/communication-bandwidth" title="Learn more about communication bandwidth from ScienceDirect's AI-generated Topic Pages">communication bandwidth</a> cost incurred in collecting flow statistics. LSF consists of two phases, finding the switch to be polled next and determining the best polling mechanism. As compared with an existing <a href="https://www.sciencedirect.com/topics/engineering/greedy-algorithm" title="Learn more about greedy algorithm from ScienceDirect's AI-generated Topic Pages">greedy algorithm</a>, simulation results show that our LSF not only runs faster but also cut down the bandwidth cost by about 30%.<br></p> | - |
dc.language | eng | - |
dc.publisher | Elsevier | - |
dc.relation.ispartof | Computer Networks | - |
dc.rights | This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. | - |
dc.subject | Critical Column First | - |
dc.subject | Flow monitoring scheme | - |
dc.subject | Lowest-cost Switch first | - |
dc.subject | Match structure | - |
dc.subject | SDN | - |
dc.title | Flow monitoring scheme design in SDN | - |
dc.type | Article | - |
dc.identifier.doi | 10.1016/j.comnet.2019.107007 | - |
dc.identifier.scopus | eid_2-s2.0-85075263872 | - |
dc.identifier.volume | 167 | - |
dc.identifier.isi | WOS:000510524600005 | - |
dc.identifier.issnl | 1389-1286 | - |