Advanced search
2 files | 1.31 MB Add to list

Fast failure recovery for in-band OpenFlow networks

Sachin Sharma (UGent) , Dimitri Staessens (UGent) , Didier Colle (UGent) , Mario Pickavet (UGent) and Piet Demeester (UGent)
Author
Organization
Abstract
In OpenFlow, control and data plane are decoupled from switches/routers. Direct programming of routers/switches is realised from one or more servers (so called controllers). In the case of an in-band OpenFlow network, the control traffic (traffic to or from the controllers) is sent on the same channel used to transport data traffic. Therefore, when a failure occurs along the data traffic path, both control and data traffic can be affected. This paper explains how failure recovery can be deployed in such a network. To achieve carrier-grade quality, the network should be able to recover from the failure within 50 ms. We apply two well-known recovery mechanisms -restoration and protection-for the control and the data traffic, and run extensive emulation experiments. The emulation results show that restoration does not allow to recover within 50 ms. Moreover, the restoration of the control traffic delays the restoration of the data traffic. The emulation results also show that protection for both control and data traffic can meet the carrier-grade recovery requirement, even in a large-scale network serving many flows.
Keywords
IBCN

Downloads

  • 5585 i.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 632.56 KB
  • (...).pdf
    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 679.85 KB

Citation

Please use this url to cite or link to this publication:

MLA
Sharma, Sachin et al. “Fast Failure Recovery for In-band OpenFlow Networks.” 9th International Conference on Design of Reliable Communication Networks, Proceedings. IEEE, 2013. 52–59. Print.
APA
Sharma, Sachin, Staessens, D., Colle, D., Pickavet, M., & Demeester, P. (2013). Fast failure recovery for in-band OpenFlow networks. 9th International Conference on Design of Reliable Communication Networks, Proceedings (pp. 52–59). Presented at the 9th International Conference on Design of Reliable Communication Networks (DRCN - 2013), IEEE.
Chicago author-date
Sharma, Sachin, Dimitri Staessens, Didier Colle, Mario Pickavet, and Piet Demeester. 2013. “Fast Failure Recovery for In-band OpenFlow Networks.” In 9th International Conference on Design of Reliable Communication Networks, Proceedings, 52–59. IEEE.
Chicago author-date (all authors)
Sharma, Sachin, Dimitri Staessens, Didier Colle, Mario Pickavet, and Piet Demeester. 2013. “Fast Failure Recovery for In-band OpenFlow Networks.” In 9th International Conference on Design of Reliable Communication Networks, Proceedings, 52–59. IEEE.
Vancouver
1.
Sharma S, Staessens D, Colle D, Pickavet M, Demeester P. Fast failure recovery for in-band OpenFlow networks. 9th International Conference on Design of Reliable Communication Networks, Proceedings. IEEE; 2013. p. 52–9.
IEEE
[1]
S. Sharma, D. Staessens, D. Colle, M. Pickavet, and P. Demeester, “Fast failure recovery for in-band OpenFlow networks,” in 9th International Conference on Design of Reliable Communication Networks, Proceedings, Budapest, Hungary, 2013, pp. 52–59.
@inproceedings{3260137,
  abstract     = {In OpenFlow, control and data plane are decoupled from switches/routers. Direct programming of routers/switches is realised from one or more servers (so called controllers). In the case of an in-band OpenFlow network, the control traffic (traffic to or from the controllers) is sent on the same channel used to transport data traffic. Therefore, when a failure occurs along the data traffic path, both control and data traffic can be affected. This paper explains how failure recovery can be deployed in such a network. To achieve carrier-grade quality, the network should be able to recover from the failure within 50 ms. We apply two well-known recovery mechanisms -restoration and protection-for the control and the data traffic, and run extensive emulation experiments. The emulation results show that restoration does not allow to recover within 50 ms. Moreover, the restoration of the control traffic delays the restoration of the data traffic. The emulation results also show that protection for both control and data traffic can meet the carrier-grade recovery requirement, even in a large-scale network serving many flows.},
  author       = {Sharma, Sachin and Staessens, Dimitri and Colle, Didier and Pickavet, Mario and Demeester, Piet},
  booktitle    = {9th International Conference on Design of Reliable Communication Networks, Proceedings},
  isbn         = {9781479900497},
  keywords     = {IBCN},
  language     = {eng},
  location     = {Budapest, Hungary},
  pages        = {52--59},
  publisher    = {IEEE},
  title        = {Fast failure recovery for in-band OpenFlow networks},
  year         = {2013},
}

Web of Science
Times cited: