Control Plane Performance Research Articles

Exploring Data Plane Updates on P4 Switches with P4Runtime

The development and roll-out of new Ethernet standards increase the available bandwidths in computer networks. This growth presents significant advantages, enabling novel applications. At the same time, the increase introduces new challenges; higher data rates reduce the available time budget to process each packet. This development also impacts software-defined networks. Their data planes need to keep up with the increased traffic rates. Nevertheless, the control plane must not be ignored; fast reaction times are necessary to handle the increased rates handled by data planes efficiently.In our work, we analyze the interaction of a high-performance data plane and different implementations for the control plane. We selected a P4 switching ASIC as our data plane. For the control plane, we investigate vendor-specific implementations and a standardized implementation called P4Runtime. To determine the performance of the control plane, we introduce a novel measurement methodology. This methodology allows measuring the delay between the initiation of rule updates on the control plane and their application on the data plane. We investigate the behavior of the data plane, its performance and non-atomicity of updates. Based on our findings, we apply different optimization strategies to improve control plane performance. Our measurements show that neglecting the control plane performance may impact network behavior due to delayed updates, but we also show how to minimize this delay and, thereby, its impact. We have released the experiment artifacts of our study including experiment scripts and measurement data.

A Robust Controller Deployment Algorithm under Multiple Constraints for SDON

The deployment of controller for SDON (Software Defined Optical Networks) is an NP-hard multi-objective optimization problem, and it is difficult to find an optimal solution that satisfies all objectives. Therefore, to improve the performance of control plane, the model we proposed is constructed based on three constraints on the survivability, delay, and control redundancy, which has three advantages. first, the survivability is improved by not sharing the risk between the main control channel and the protection control channel, which also improves the utilization of network resources. Second, the accuracy of delay calculation is improved by using a delay optimization model, and a threshold value is set to constrain the delay. Finally, the greedy strategy idea is used to achieve the full network coverage of the control range and improve the convergence speed. Simulation results show that the algorithm improves the fault recovery probability by an average of 15% compared with the algorithm of KSP computed routing. In addition, the algorithm in this paper achieves the same survivability as the C-MPC algorithm while using an average 11.1% reduction in the number of controllers.

Control Plane Performance Benchmarking and Feature Analysis of Popular Open-Source 5G Core Networks: OpenAirInterface, Open5GS, and free5GC

Control Plane Performance Benchmarking and Feature Analysis of Popular Open-Source 5G Core Networks: OpenAirInterface, Open5GS, and free5GC

Quality of transmission-aware control plane performance analysis for elastic optical networks

Physical layer impairment-aware control plane techniques are studied for dynamic networking of elastic optical networks (EONs). To this end, performance of three control plane architectures, the generalized multi-protocol label switching (GMPLS), the GMPLS method with a path computation element unit, and the software-defined networking (SDN), are analyzed considering nonlinear interference (NLI) noise (consisting of self-phase and cross-phase modulation effects) and amplified spontaneous emission noise. Signal-to-noise ratio estimation of all the control plane architectures is performed by means of the well established gaussian noise model. To reduce the blocking probability caused by the NLI noise in EONs, an impairment-aware routing and spectrum assignment algorithm is proposed to ensure a required quality of transmission for requested light-paths. Moreover, transmit launch power influence on blocking probability is investigated exploiting extensive numerical simulations based on stochastic traffic modeling for the investigated control plane architectures in Pan-European and national science foundation network (NSFNET) topologies. The numerical results show nearly 50% improvement in blocking probability of the SDN control plane by applying an impairment aware scheme at traffics around 200 Erlang. Furthermore, the results indicate that the lowest blocking probability is observed at a transmit power of 0 dBm for all the control plane techniques.

SAHAR: A Control Plane Architecture for High Available Software-Defined Networks

Software-defined networking (SDN) is an architectural paradigm which enables a centralised control by decoupling the data and control planes of a network. In reactive mode of SDN, the significant variation in flow arrival events places considerable stress on the control plane. Also, the existence of frequent events such as network-wide statistics collection which significantly interfere with the basic functions of control plane can drastically affect performance of control plane. Therefore, this paper introduces a new architectural model named SAHAR that uses a controller box consisting of a coordinator controller and a primary flow-setup controller and one or more secondary flow-setup controllers as needed. Assigning monitor and management tasks to the coordinator controller reduces the load on the flowsetup controllers. Also, dividing input traffic between flow-setup controllers by the coordinator controller distributes the load in the control plane. In this way, SAHAR achieves a high available SDN which can be scalable.

Reducing controller response time with hybrid routing in software defined networks

In a typical software defined network (SDN), switches report the Packet-In messages of newly arrived flows to the controllers. With more and more flows arriving at a network, the controller load significantly increases, which may lead to long (or even unacceptable) controller response time. Though previous solutions, such as dynamic controller assignment, help to reduce the controller response time, they still lead to various disadvantages, such as an unacceptable controller re-assignment delay, massive communication overhead between controllers, or poor routing performance. To address this issue, our intention is to trade data plane resource for better control plane performance. Specifically, we propose to pre-install wildcard entries for some aggregated flows to reduce controller response time, and perform dynamic routing for new-arrival flows to optimize the network performance. We define the problem of reducing controller response time with minimum data plane resource cost, and prove its NP-hardness. We then present an efficient algorithm based on randomized rounding, and analyze that our algorithm can achieve constant bicriteria approximation under most practical situations. Some practical issues are discussed to enhance our algorithm. We have implemented the proposed algorithm on a real SDN testbed. The experimental results and the extensive simulation results show that our method can reduce the controller response time by 47%, or improve the network throughput by 56% compared with the previous solutions, even with significant traffic dynamics.

Enhancing the Isolation and Performance of Control Planes for Fog Computing.

Fog computing, which places computing resources close to IoT devices, can offer low latency data processing for IoT applications. With software-defined networking (SDN), fog computing can enable network control logics to become programmable and run on a decoupled control plane, rather than on a physical switch. Therefore, network switches are controlled via the control plane. However, existing control planes have limitations in providing isolation and high performance, which are crucial to support multi-tenancy and scalability in fog computing. In this paper, we present optimization techniques for Linux to provide isolation and high performance for the control plane of SDN. The new techniques are (1) separate execution environment (SE2), which separates the execution environments between multiple control planes, and (2) separate packet processing (SP2), which reduces the complexity of the existing network stack in Linux. We evaluate the proposed techniques on commodity hardware and show that the maximum performance of a control plane increases by four times compared to the native Linux while providing strong isolation.

Deterministic Quality of Service Guarantee for Dynamic Service Chaining in Software Defined Networking

In this paper, we present a systemic approach to provide deterministic delay guarantee for dynamic service chaining in software defined networking (SDN). The delay performance of service chaining in SDN is affected by signaling message exchange in control plane and packet transmissions in data plane, respectively. First, we develop an analytical method to characterize the delay performance of control plane when handling traffic with different priorities according to network calculus (NC) and queuing theory. Second, taking into account the estimated delay in control plane, we propose a novel service traversal mechanism to calculate the optimal traversal path for the service chain. We demonstrate that NC delay analysis can provide deterministic quality of service (QoS)-guaranteed service chaining for any specified delay requirements, whereas theoretical queueing delay analysis can only provide statistical QoS guarantee. In summary, the proposed NC delay analysis can help to understand the network design for a future delay sensitive Internet in which deterministic latency must be guaranteed.

ONVisor: Towards a scalable and flexible SDN‐based network virtualization platform on ONOS

SummaryNetwork virtualization (NV) technologies have attracted a lot of attention as an essential solution for future networking infrastructure. The NV enables multiple tenants to share the same physical infrastructure and to create independent virtual networks (VNs) by decoupling the physical network in terms of topology, address, and control functions. One feasible way to realize full NV involves considering solutions based on the software‐defined networking (SDN) paradigm using its programmability. The SDN contributes many benefits to both network operations and management including programmability, agility, elasticity, and flexibility. There are several SDN‐based NV solutions; however, they suffered from a lack of scalability, high availability. Also, they have high latency between control and data plane because of proxy‐based architecture. In this thesis, we introduce a new NV platform, named Open Network Hypervisor (ONVisor). The design objectives include, among the features, (1) multitenancy, (2) scalability, (3) flexibility, (4) isolated VNs, and (5) VN federation. ONVisor was designed and implemented by extending Open Network Operating System, an open‐source SDN controller. The main features of ONVisor are (1) isolated control and data plane per VN, (2) support of distributed operations, (3) extensible translators, (4) on‐platform VN application development and execution, and (5) support of heterogenous SDN data‐plane implementations. Several experiments are conducted on various test scenarios in different test environments in terms of control and data plane performance compared to nonvirtualized SDN network. The results show that ONVisor can provide VNs a little bit lower control plane performance and similar data plane performance.

The controller placement problem or the controller selection problem?

In SDN, the control logic of packet processing devices is moved onto the SDN centralized controller. This decoupled networking architecture creates some critical concerns when compared to traditional distributed architectures. One primary concern in this architecture is the placement of the controller, which is commonly known as the CPP. From our extensive literature review, we identify that there is no strict placement rule that best fits every network in SDN. Dynamic addition and deletion of controllers is inevitable. Motivated from this, in this paper, we attempt to transform the CPP into a CSP. We show that, in the dynamic SDN environment, researchers need to pay attention to CSP issues. The clear advantage of CSP over CPP is that, in the dynamic SDN ecosystem, we can answer fundamental questions about the control plane performance, i.e., minimum number of controllers, their workload distribution, and placement/locations. Furthermore, we can meet strict, application-specific, QoS constraints.

Dynamic switch migration towards a scalable SDN control plane

SummaryDistributed control plane is a promising approach to scalable software‐defined networking (SDN). Live migration of switches from controllers that are overloaded to those that are underutilized may be a solution to handle peak switch traffic using available control resource. However, such migration has to be performed with a well‐designed mechanism to fully utilize available resources in all three resource dimensions: CPU, bandwidth, and memory. In this article, we first provide a resource model for SDN and reduce the switch migration decision to a centralized resource utilization maximization problem under constraints of CPU, bandwidth, and memory. Second, we show that the problem of maximizing resource utilization in an SDN is equivalent to that of maximizing game players' profits in the context of non‐cooperative game theory. Taking controllers and switches as game players and commodities respectively, the player policy is how to migrate switches among the control plane. Finally, we implement a proof of concept, called GAME‐Switch Migration (GAME‐SM). The numerical experiments using Mininet emulator validate nice properties of our game model in enhancing the performance of control plane in SDN. Copyright © 2016 John Wiley & Sons, Ltd.

Sparse Control Plane Design for Optical WDM Mesh Networks

In GMPLS-based WDM mesh networks, control plane differs data plane in traffic distribution, QoS, and survivability requirements. Usually we use wavelengths from data plane to construct control channels. The idea of adding routing capability into control plane has been proposed recently to enhance survivability of control plane. In this paper, by taking advantage of routing capability of control plane, we motivate the needs of designing control plane with different topology with data plane, i.e., sparse control plane, so as to save wavelengths and reduce cost. A heuristic algorithm called Adaptive Optimal Sparse Topology Discovery (AOSTD) is proposed. Analytical and simulation results demonstrate that our proposed algorithm can simplify control plane topology without compromising the performance of control plane. The proposed algorithm has the advantages of fast convergence and adaptability.