SDN Controller Placement at the Edge: Optimizing Delay and Overheads

Qin, Qiaofeng; Poularakis, Konstantinos; Iosifidis, George; Tassiulas, Leandros

doi:10.1109/infocom.2018.8485963

Cited by 61 publications

(30 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As discussed briefly in Section I, frequent dissemination of inter-controller synchronization messages introduce prohibitively large overheads [3], [4]. Moreover, excessive status updates can potentially lead to network performance degradation caused by instability [14].…”

Section: B Synchronization Among Sdn Controllersmentioning

confidence: 99%

See 1 more Smart Citation

MACS: Deep Reinforcement Learning based SDN Controller Synchronization Policy Design

Zhang

Poularakis

et al. 2019

2019 IEEE 27th International Conference on Network Protocols (ICNP)

Self Cite

View full text Add to dashboard Cite

In distributed software-defined networks (SDN), multiple physical SDN controllers, each managing a network domain, are implemented to balance centralised control, scalability, and reliability requirements. In such networking paradigms, controllers synchronize with each other, in attempts to maintain a logically centralised network view. Despite the presence of various design proposals for distributed SDN controller architectures, most existing works only aim at eliminating anomalies arising from the inconsistencies in different controllers' network views. However, the performance aspect of controller synchronization designs with respect to given SDN applications are generally missing. To fill this gap, we formulate the controller synchronization problem as a Markov decision process (MDP) and apply reinforcement learning techniques combined with deep neural networks (DNNs) to train a smart, scalable, and fine-grained controller synchronization policy, called the Multi-Armed Cooperative Synchronization (MACS), whose goal is to maximise the performance enhancements brought by controller synchronizations. Evaluation results confirm the DNN's exceptional ability in abstracting latent patterns in the distributed SDN environment, rendering significant superiority to MACS-based synchronization policy, which are 56% and 30% performance improvements over ONOS and greedy SDN controller synchronization heuristics.

show abstract

Section: B Synchronization Among Sdn Controllersmentioning

confidence: 99%

“…. , N } 4. while t ≤ T do = 0, 1 with probabilities 1 − , , respectively; an action at according to Section IV-D; 11 end 12…”

mentioning

confidence: 99%

MACS: Deep Reinforcement Learning based SDN Controller Synchronization Policy Design

Zhang

Poularakis

et al. 2019

2019 IEEE 27th International Conference on Network Protocols (ICNP)

Self Cite

View full text Add to dashboard Cite

show abstract

“…In such a case, the standard problem of submodularity will attempt to find an optimal subset of nodes that may not contain pre-defined nodes. This use case and further possible scenarios in SDN controller placements cannot be simply considered [13]. Therefore, in this work, we propose a robust framework to address varying use cases of this problem.…”

Section: Multiple Placement Schemementioning

confidence: 99%

SDN Controller Placement in IoT Networks: An Optimized Submodularity-Based Approach

Khoa

Piran

Pham

2019

Sensors

View full text Add to dashboard Cite

Software-Defined Networking (SDN) has opened a promising and potential approach for future networks, which mostly requires the low-level configuration to implement different controls. With the high advantages of SDN by decomposing the network control plane from the data plane, SDN has become a crucial platform to implement Internet of Things (IoT) services. However, a static SDN controller placement cannot obtain an efficient solution in distributed and dynamic IoT networks. In this paper, we investigate an optimization framework under a well-known theory, namely submodularity optimization, to formulate and address different aspects of the controller placement problem in a distributed network, specifically in an IoT scenario. Concretely, we develop a framework that deals with a series of controller placement problems from basic to complicated use cases. Corresponding to each use case, we provide discussion and a heuristic algorithm based on the submodularity concept. Finally, we present extensive simulations conducted on our framework. The simulation results show that our proposed algorithms can outperform considered baseline methods in terms of execution time, the number of controllers, and network latency.

show abstract

“…. , τ }; 3 Estimate Ψ( x) = 1 τ τ t=1 Ψ t (x t ); 4 for each iteration k from 1 to B do 5 Pick σ random controller pairs p for which x p < R; 6 for each picked pair p from 1 to σ do 7 Set x = x where x p = x p + 1; 8 Try out x t = x and observe Ψ t (x t ) ∀t ∈ {(k−1)στ +pτ +1, . .…”

Section: Learning Algorithmmentioning

confidence: 99%

“…Distributed SDN controller deployments require a coordination protocol among controllers, which could easily generate significant amount of control traffic, e.g., see the measurement studies in [6] and [7]. However, the control traffic is often neglected in literature with most of the existing works focusing on the routing and balancing of the data traffic, e.g., see [16] and the survey in [1].…”

Section: Related Workmentioning

confidence: 99%

Learning the Optimal Synchronization Rates in Distributed SDN Control Architectures

Poularakis

Qin

et al. 2019

IEEE INFOCOM 2019 - IEEE Conference on Computer Communications

Self Cite

View full text Add to dashboard Cite

Since the early development of Software-Defined Network (SDN) technology, researchers have been concerned with the idea of physical distribution of the control plane to address scalability and reliability challenges of centralized designs. However, having multiple controllers managing the network while maintaining a "logically-centralized" network view brings additional challenges. One such challenge is how to coordinate the management decisions made by the controllers which is usually achieved by disseminating synchronization messages in a peer-to-peer manner. While there exist many architectures and protocols to ensure synchronized network views and drive coordination among controllers, there is no systematic methodology for deciding the optimal frequency (or rate) of message dissemination. In this paper, we fill this gap by introducing the SDN synchronization problem: how often to synchronize the network views for each controller pair. We consider two different objectives; first, the maximization of the number of controller pairs that are synchronized, and second, the maximization of the performance of applications of interest which may be affected by the synchronization rate. Using techniques from knapsack optimization and learning theory, we derive algorithms with provable performance guarantees for each objective. Evaluation results demonstrate significant benefits over baseline schemes that synchronize all controller pairs at equal rate.

show abstract

SDN Controller Placement at the Edge: Optimizing Delay and Overheads

Cited by 61 publications

References 20 publications

MACS: Deep Reinforcement Learning based SDN Controller Synchronization Policy Design

MACS: Deep Reinforcement Learning based SDN Controller Synchronization Policy Design

SDN Controller Placement in IoT Networks: An Optimized Submodularity-Based Approach

Learning the Optimal Synchronization Rates in Distributed SDN Control Architectures

Contact Info

Product

Resources

About