Brandon Heller
Nick McKeown
Abstract
Network architectures such as Software-Defined Networks (SDNs) move the control logic off packet processing devices and onto external controllers. These network architectures with decoupled control planes open many unanswered questions regarding reliability, scalability, and performance when compared to more traditional purely distributed systems. This paper opens the investigation by focusing on two specific questions: given a topology, how many controllers are needed, and where should they go? To answer these questions, we examine fundamental limits to control plane propagation latency on an upcoming Internet2 production deployment, then expand our scope to over 100 publicly available WAN topologies. As expected, the answers depend on the topology. More surprisingly, one controller location is often sufficient to meet existing reaction-time requirements (though certainly not fault tolerance requirements).
- Ansi t1.tr.68-2001 enhanced network survivability performance.Google Scholar
- BGP Route Reflection: An Alternative to Full Mesh Internal BGP (IBGP). http://tools.ietf.org/html/rfc4456.Google Scholar
- Control And Provisioning of Wireless Access Points (CAPWAP) Protocol Specification. http://tools.ietf.org/html/rfc5415.Google Scholar
- Internet2 open science, scholarship and services exchange. http://www.internet2.edu/network/ose/.Google Scholar
- Path Computation Clients (PCC) - Path Computation Element (PCE) Requirements for Point-to-Multipoint MPLS-TE. http://tools.ietf.org/html/rfc5862.Google Scholar
- V. Arya, N. Garg, R. Khandekar, A. Meyerson, K. Munagala, and V. Pandit. Local search heuristics for k-median and facility location problems. SIAM Journal on Computing, 33(3):544--562, 2004. Google ScholarDigital Library
- M. Caesar, D. Caldwell, N. Feamster, J. Rexford, A. Shaikh, and J. van der Merwe. Design and implementation of a routing control platform. In NSDI. USENIX, 2005. Google ScholarDigital Library
- M. Casado, M. Freedman, J. Pettit, J. Luo, N. McKeown, and S. Shenker. Ethane: Taking control of the enterprise. ACM SIGCOMM CCR, 37(4):1--12, 2007. Google ScholarDigital Library
- A. Greenberg, G. Hjalmtysson, and et al. A clean slate 4D approach to network control and management. ACM SIGCOMM CCR, 35(5):54, 2005. Google ScholarDigital Library
- T. Griffin and G. Huston. BGP Wedgies. RFC 4264 (Informational), Nov. 2005.Google Scholar
- T. G. Griffin and G. Wilfong. An analysis of bgp convergence properties. SIGCOMM CCR., 29:277--288, August 1999. Google ScholarDigital Library
- N. Gude, T. Koponen, J. Pettit, B. Pfaff, M. Casado, and N. McKeown. Nox: Towards an operating system for networks. In ACM SIGCOMM CCR, July 2008. Google ScholarDigital Library
- N. Handigol, S. Seetharaman, M. Flajslik, R. Johari, and N. McKeown. Aster*x: Load-balancing as a network primitive. 9th GENI Engineering Conference (Plenary), November 2010.Google Scholar
- D. Hochba. Approximation algorithms for np-hard problems. ACM SIGACT News, 28(2):40--52, 1997. Google ScholarDigital Library
- S. Knight, H. Nguyen, N. Falkner, R. Bowden, and M. Roughan. The internet topology zoo.Google Scholar
- T. Koponen, M. Casado, and et al. Onix: A distributed control platform for large-scale production networks. In OSDI. USENIX, 2010. Google ScholarDigital Library
- Z. M. Mao, R. Govindan, G. Varghese, and R. H. Katz. Route flap damping exacerbates internet routing convergence. SIGCOMM CCR, 32:221--233, August 2002. Google ScholarDigital Library
- The openflow switch. http://www.openflowswitch.org.Google Scholar
- R. Sherwood, G. Gibb, K.-K. Yap, G. Appenzeller, M. Casado, N. McKeown, and G. Parulkar. Can the Production Network Be the Testbed? In OSDI. USENIX, 2010. Google ScholarDigital Library
- M. Shindler. Approximation algorithms for the metric k-median problem. Written Qualifying Exam Paper, University of California, Los Angeles. Cited on, page 44.Google Scholar
- V. Vazirani. Approximation algorithms. Springer Verlag, 2001. Google ScholarDigital Library
- H. Yan, D. Maltz, T. Ng, H. Gogineni, H. Zhang, and Z. Cai. Tesseract: A 4d network control plane. In NSDI. USENIX, 2007. Google ScholarDigital Library
Index Terms
- The controller placement problem
Recommendations
The controller placement problem
HotSDN '12: Proceedings of the first workshop on Hot topics in software defined networksNetwork architectures such as Software-Defined Networks (SDNs) move the control logic off packet processing devices and onto external controllers. These network architectures with decoupled control planes open many unanswered questions regarding ...
Density cluster based approach for controller placement problem in large-scale software defined networkings
Software Defined Networking (SDN) decouples control and data planes. The separation arises a problem known as the controller placement, i.e., how many and where controllers should be deployed. Currently, most works defined this problem as the multi-...
Varna-based optimization: a novel method for capacitated controller placement problem in SDN
AbstractRecently, software defined networking (SDN) is a promising paradigm shift that decouples the control plane from the data plane. It can centrally monitor and control the network through softwarization, i.e., controller. Multiple controllers are a ...
Comments