OpenFlow Switching: Data Plane Performance

Bianco, A.; Birke, Robert; Giraudo, Luca; Palacin, M.

doi:10.1109/icc.2010.5502016

Cited by 126 publications

(75 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The OpenFlow protocol reference is the software switch, OpenVSwitch [17], an important implementation for production environments. Firstly, OpenVSwitch provides a replacement for the poor-performing Linux bridge [7], a crucial functionality for virtualised operating systems. Secondly, several hardware switch vendors use OpenVSwitch as the basis for the development of their own OpenFlow-enabled firmware.…”

Section: Measurement Setupmentioning

confidence: 99%

“…Yet, there have been few performance studies: to our knowledge, OFLOPS is the first attempt to develop a platform that is able to provide detailed measurements for the OpenFlow implementations. Bianco et al [7] show the performance advantage of the Linux software OpenFlow over the Linux Ethernet switch, while Curtis et al [9] discuss some design limitations of the protocol when deployed in large network environments. We consider OFLOPS, alongside [10], as one of a new generation of measurement systems that, like the intelligent traffic and router evaluators [13,4], go beyond simple packet-capture.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

OFLOPS: An Open Framework for OpenFlow Switch Evaluation

Rotsos

Sarrar

Uhlig

et al. 2012

Lecture Notes in Computer Science

224

134

View full text Add to dashboard Cite

Abstract. Recent efforts in software-defined networks, such as OpenFlow, give unprecedented access into the forwarding plane of networking equipment. When building a network based on OpenFlow however, one must take into account the performance characteristics of particular OpenFlow switch implementations. In this paper, we present OFLOPS, an open and generic software framework that permits the development of tests for OpenFlow-enabled switches, that measure the capabilities and bottlenecks between the forwarding engine of the switch and the remote control application. OFLOPS combines hardware instrumentation with an extensible software framework. We use OFLOPS to evaluate current OpenFlow switch implementations and make the following observations: (i) The switching performance of flows depends on applied actions and firmware. (ii) Current OpenFlow implementations differ substantially in flow updating rates as well as traffic monitoring capabilities. (iii) Accurate OpenFlow command completion can be observed only through the data plane. These observations are crucial for understanding the applicability of OpenFlow in the context of specific use-cases, which have requirements in terms of forwarding table consistency, flow setup latency, flow space granularity, packet modification types, and/or traffic monitoring abilities.

show abstract

Section: Measurement Setupmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

OFLOPS: An Open Framework for OpenFlow Switch Evaluation

Rotsos

Sarrar

Uhlig

et al. 2012

Lecture Notes in Computer Science

224

134

View full text Add to dashboard Cite

show abstract

“…However, the model is depicted only for a single node in the data plane and the time stopping method employed therein has limited real time application. Secondly the framework is based on deterministic network calculus which does not provide any meaningful bounds (Bianco and Birke, 2010).…”

Section: Related Workmentioning

confidence: 99%

A Framework of Controller with Flow Table Cache and Performance Analysis in Software Defined Industrial Networks

2016

Rev. Téc. Ing. Univ. Zulia

View full text Add to dashboard Cite

The industrial internet and industrial 4.0 introduce the new requirements of industrial networks to be more flexible, manageable and scalable. Software Defined Network (SDN) separates control plane and data plane, abstracts the complexity of the network, provides faster network services and more flexible network management, and bring great impact to the traditional industrial networks. However, due to the limit of power and cost in industrial network switches, the scale of flow table of these switches is smaller than the scale of switches in datacenter, which leads to a low flow table hit ratio and heavy load in SDN controller. In this paper, we propose a framework of SDN controller with flow table cache and increase the hit ratio. Based on the framework, a model of SDN controller is proposed and the performance of SDN controller can be analyzed. The simulation shows the analysis results from the model is close to the simulation and the controller with flow table cache is more efficient in software defined industrial networks.

show abstract

“…Failing these, however, brings undue cache thrashing and packets recurring to the slow path. This is then perceived by the user as perplexing throughput drops, latency spikes, and downright service interruptions, on various hard-to-predict combinations of flow tables and input traffic [29][30][31][32][33][34]. Vexing cache management complexity.…”

Section: The Case Against Flow Cachingmentioning

confidence: 99%

“…It is well-known, however, that flow caching performs poorly for many important applications that require forwarding decisions depending on diverse "high-entropy" packet fields, like transportlayer firewalls, or produce short-lived flows, e.g., peer-topeer protocols, MapReduce, or network monitoring [19,29]. Consequently, OpenFlow switches often exhibit abrupt performance regressions in various hard-to-predict combinations of flow tables and traffic patterns [29][30][31][32][33][34], opening the door to malicious denial-of-service-like attacks even on as innocently looking traffic patterns as port scans [19,29,35].…”

Section: Introductionmentioning

confidence: 99%

Dataplane Specialization for High-performance OpenFlow Software Switching

Molnár

Pongrácz

Enyedi

et al. 2016

Proceedings of the 2016 ACM SIGCOMM Conference

View full text Add to dashboard Cite

OpenFlow is an amazingly expressive dataplane programming language, but this expressiveness comes at a severe performance price as switches must do excessive packet classification in the fast path. The prevalent OpenFlow software switch architecture is therefore built on flow caching, but this imposes intricate limitations on the workloads that can be supported efficiently and may even open the door to malicious cache overflow attacks. In this paper we argue that instead of enforcing the same universal flow cache semantics to all OpenFlow applications and optimize for the common case, a switch should rather automatically specialize its dataplane piecemeal with respect to the configured workload. We introduce ESWITCH, a novel switch architecture that uses on-the-fly template-based code generation to compile any OpenFlow pipeline into efficient machine code, which can then be readily used as fast path. We present a proofof-concept prototype and we demonstrate on illustrative use cases that ESWITCH yields a simpler architecture, superior packet processing speed, improved latency and CPU scalability, and predictable performance. Our prototype can easily scale beyond 100 Gbps on a single Intel blade even with complex OpenFlow pipelines.

show abstract

OpenFlow Switching: Data Plane Performance

Cited by 126 publications

References 3 publications

OFLOPS: An Open Framework for OpenFlow Switch Evaluation

OFLOPS: An Open Framework for OpenFlow Switch Evaluation

A Framework of Controller with Flow Table Cache and Performance Analysis in Software Defined Industrial Networks

Dataplane Specialization for High-performance OpenFlow Software Switching

Contact Info

Product

Resources

About