Are Today's SDN Controllers Ready for Primetime?

Mallon, Stephen; Gramoli, Vincent; Jourjon, Guillaume

doi:10.1109/lcn.2016.60

Cited by 12 publications

(5 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This paper does however focus on the "black box approach" trying to distinguish limitations of any controllers, no matter the inner logic that is behind their workings. Papers that provide further insightby comparing controllersinto this subject are already available to the research community [22], [28].…”

Section: ) Controller's Objectivesmentioning

confidence: 99%

Impact of SDN Controller’s Performance on Quality of Service

Kaczmarek,

Litka

2024

IEEE Access

View full text Add to dashboard Cite

Software Defined Networking is a paradigm in network architecture; that is quickly becoming commonplace in modern telecommunication systems. It facilitates network customization for the requirements of different applications and simplifies the implementation of new services. Since its proposal, a significant evolution in its functionality has occurred. However, this development brought along problems of efficiency and performance, which are currently under research. A number of requirements has to be met, if Software Defined Networking is going to be the next step in the Next-Generation Networks progression. The central part of itthe SDN controllerhas to put minimum strain on the system and provide performance which does not impede Quality of Service requirements. In this paper, the results of a research on SDN controller's performance have been provided in the context of keeping up with flows' QoS. For this, an emulated-physical research platform has been implemented. This research environment utilizes traffic generated accordingly to ITU-T recommendations to validate QoS parameters. The platform is given a thorough description. The results obtained from it take under consideration the implemented traffic sources, as well as the intensity of traffic handled by the controller and the traffic load of data plane links. Authors indicate that even without breaking the limitations set for delays in QoS, the impact of the controller workload is noticeable, which should be mitigated by applying appropriate resource control.

show abstract

Section: ) Controller's Objectivesmentioning

confidence: 99%

Impact of SDN Controller’s Performance on Quality of Service

Kaczmarek,

Litka

2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The only related work in terms of benchmarking builds on top of OF controllers. Due to the maturity of OF, there is indeed abundant literature in this regard, e.g., [2], [14]- [21]. The main focus is to evaluate the performance in terms of latency and throughput, and it has been concluded that the performance highly depends on the threading capability of the controller, which shows the number of requests it can process in a single time slot [2].…”

Section: Related Workmentioning

confidence: 99%

“…The main focus is to evaluate the performance in terms of latency and throughput, and it has been concluded that the performance highly depends on the threading capability of the controller, which shows the number of requests it can process in a single time slot [2]. In addition, reliability and energy consumption are evaluated in some works [16], [21].…”

Section: Related Workmentioning

confidence: 99%

P4RCProbe for Evaluating the Performance of P4Runtime-based Controllers

Harkous

Sherkawi

Jarschel

et al. 2021

2021 IEEE Conference on Network Function Virtualization and Software Defined Networks (NFV-SDN)

View full text Add to dashboard Cite

The SDN control plane provides the data plane with packet forwarding rules and thus defines its behavior. Understanding the performance of the controller is critical to assess the overall performance of the network. Although OpenFlow-based controllers have been well investigated in this regard, controller performance studies based on P4Runtime, which is the de-facto implementation for the southbound control of programmable data planes in P4, are still missing.In this work, we implement a benchmarking tool for P4Runtime-based controllers, apply the tool to evaluate the performance of the ONOS controller running in both OpenFlow and P4Runtime mode, and identify processing bottlenecks in the P4Runtime implementation. In addition, We propose a code patch for the implementation, which shows a 17 % improvement in the achieved packet rate.

show abstract

“…Studies of data centre traffic characteristics [18, 48] have shown that 99% of flows in a data centre are ‘mice’ flows (i.e. a few megabytes in size) and are completed within 100 ms. An evaluation of the performance of controllers [49] has shown that the best possible response time for a reasonable workload is more than 10 ms, which is more than 10% overhead for 99% of the flows. Such a high overhead is unacceptable for many cloud applications.…”

Section: Overview Of the Dcnet Network Architecturementioning

confidence: 99%

DCnet: a data centre network architecture that supports live VM migration

Comer

Karandikar

2019

IET Networks

View full text Add to dashboard Cite

Cloud computing and storage have become an integral part of online Internet services. Over the past decade, enterprises have steadily moved their applications and storage to the cloud paradigm. Server virtualisation enables cloud providers to utilise computational resources efficiently by employing virtual machines (VMs) and containers. As more enterprise applications are hosted by VMs in the cloud, a significant amount of network communication occurs among servers within a data centre, increasing so-called 'east-west communication'. In addition to scaling to accommodate more east-west traffic, future data centre networks must support efficient VM migration, which is used to optimise power consumption. This article proposes a new data centre network architecture called DCnet. DCnet changes addressing and routing at layers 2 and 3 completely to increase throughput and support live VM migration throughout an organisation, including across multiple data centres owned by an organisation that spans large geographic areas. DCnet retains compatibility with existing hardware by using the same frame format as standard protocols, specifically Ethernet and Internet protocol (IP). Furthermore, DCnet does not require any changes to applications and host operating systems or libraries. In addition to presenting the architecture, the article describes a test bed used to assess DCnet and reports experimental measurements.

show abstract

Are Today's SDN Controllers Ready for Primetime?

Cited by 12 publications

References 11 publications

Impact of SDN Controller’s Performance on Quality of Service

Impact of SDN Controller’s Performance on Quality of Service

P4RCProbe for Evaluating the Performance of P4Runtime-based Controllers

DCnet: a data centre network architecture that supports live VM migration

Contact Info

Product

Resources

About