Flow Table Management Research Articles

Flow-based Service Time optimization in software-defined networks using Deep Reinforcement Learning

A Ternary Content-Addressable Memory (TCAM) constitutes a memory variant employed within Software Defined Networking (SDN) node flow tables. These TCAMs deliver swift processing, enabling rapid parallel lookups. Nonetheless, due to their high energy consumption and cost, TCAMs have limited dimensions. This size constraint influences rule capacity, and suboptimal rule management can degrade network quality of service. Although different techniques for flow table management have been proposed during recent years, such as eviction, idle and hard timeout mechanisms, this paper proposes a Deep-Reinforcement Learning (DRL) solution, namely DRL-Idle, that is able to maximize the service time of the flows in an SDN network without considering any assumption about their status throughout time. By means of a continuous learning, DRL-Idle is able to also minimize the number of rule installations that are required to serve the target flows. Based on the idea of dynamically modifying the idle timeout value of the flows according to their needs, DRL-Idle outperforms existing approaches aiming at solving the Rule Placement Problem, averaging a 30% increase in performance.

A two-phase detection method against APT attack on flow table management in SDN

A two-phase detection method against APT attack on flow table management in SDN

Software Defined Networking Flow Table Management of OpenFlow Switches Performance and Security Challenges: A Survey

Software defined networking (SDN) is an emerging network paradigm that decouples the control plane from the data plane. The data plane is composed of forwarding elements called switches and the control plane is composed of controllers. SDN is gaining popularity from industry and academics due to its advantages such as centralized, flexible, and programmable network management. The increasing number of traffics due to the proliferation of the Internet of Thing (IoT) devices may result in two problems: (1) increased processing load of the controller, and (2) insufficient space in the switches’ flow table to accommodate the flow entries. These problems may cause undesired network behavior and unstable network performance, especially in large-scale networks. Many solutions have been proposed to improve the management of the flow table, reducing controller processing load, and mitigating security threats and vulnerabilities on the controllers and switches. This paper provides comprehensive surveys of existing schemes to ensure SDN meets the quality of service (QoS) demands of various applications and cloud services. Finally, potential future research directions are identified and discussed such as management of flow table using machine learning.

AdaptFlow: Adaptive Flow Forwarding Scheme for Software-Defined Industrial Networks

Industry 4.0 revolution has emerged as an escalator for increased productivity and cost savings in smart factories. However, they pose a big challenge for network architecture in providing a flexible way for continuous data flow. The necessity of reliable connectivity in the industrial ecosystem has paved the path for the evolution of software-defined industrial networks. However, the nature of data traffic in the industrial environment further pushes the need for adaptive flow forwarding and control. Therefore, in this article, AdaptFlow, an adaptive flow forwarding scheme, is designed for software-defined industrial networks. AdaptFlow includes: 1) application-specific traffic classification using self-organized maps; 2) B+ tree-based flow table management; 3) queuing model for analyzing the waiting time; and 4) energy-aware flow forwarding algorithm. The evaluation of AdaptFlow scheme shows improvements in minimizing delay and energy consumption while increasing network capacity in contrast to the OpenFlow architecture.

Flow-table updating strategy for efficient use of renewable energy in software defined wireless relay networks

Software defined networking (SDN) is revolutionizing the telecommunication networking industry by providing flexible and efficient management. Because of energy cost and environmental concerns, green networking technologies are being developed to increasing energy efficiency and making use of renewable energy, for both wired and wireless networks. Relay nodes (RNs) are widely used in wireless networks to increasing energy efficiency as well as the network coverage. In this paper, we consider an SDN-enabled wireless relay network, in which the relay nodes are powered by renewable energy. In the beginning, all the users can connect to a BS or an RN in the most energy efficient way, however, the availability of renewable energy is unpredictable, and users connected to an RN will have to be handed over to base stations (BSs) when there is not enough renewable energy to power the RN. There are also occasions when an RN harvested more renewable energy than required to server the users connected to it, in these cases users connected to BS or other RNs can be transferred to connect through this RN, otherwise the harvested surplus renewable energy will be wasted as it cannot be stored. We transform the problem of optimizing the use of harvested renewable energy into a flow-table management problem. A model of harvested renewable energy available for RNs is built first, then we propose a dynamic flow-table updating strategy (DFTUS), which is driven by the availability of renewable energy, to maximize the use of the harvested renewable energy and thus minimize the use of traditional energy. DFTUS can be integrated as an application on top of an SDN controller. Simulation results show that DFTUS can make more efficient use of renewable energy thus make the overall relay network more energy efficient compared with other routing strategies such as the capacity limited strategy (CLS) and the optimal path strategy (OPS).

SAT-FLOW: Multi-Strategy Flow Table Management for Software Defined Satellite Networks

Software-defined satellite network (SDSN) is a novel framework, which brings softwaredefined network technologies in the satellite networks. It has great potential to achieve effective and flexible management in the satellite networks. There are two burning issues to be solved for the flow table management in SDSN. First, the ternary content addressable memory (TCAM) space is limited on satellites and the flow table size should be reduced. Second, the frequent handovers will lead to an increase in the flow table size in SDSN. Due to the limited flow table space, a lot of flows will be dropped if the flow table is full during the handover. To address these issues, we first give a description of our focused flow table management problems. Then, we propose SAT-FLOW, a multi-strategy flow table management method for SDSN. SAT-FLOW considers three key points, limited TCAM space, classified traffic, and handover. SAT-FLOW contains two heuristic algorithms, named dynamic classified timeout (DCT) algorithm and timeout strategybased mobility management (TSMM) algorithm. DCT aims to reduce the flow table size and TSMM aims to reduce the drop flows during the handover. We implement SAT-FLOW and conduct contrast experiments. The experimental results verify the good performance in terms of transmission quality, idle_timeout values distribution, a 15.27%-24.34% decrease in flow table size, an 8.2%-10.4% decrease in drop-flow rate, and a 4.92%-5.7% decrease in table misses for the high priority traffic during the handover.

Enhanced Flow Table Management Scheme With an LRU-Based Caching Algorithm for SDN

We propose a flow table management scheme for OpenFlow switches to minimize table-miss in a flow table. Commercial OpenFlow switches have one or more flow tables that consist of flow entries for processing packets. However, flow entries are managed based only on their timeout parameters, meaning they are expired and deleted by switches regardless of their reusability. The absence of the flow entry then causes table-misses in the future. We propose a solution for the problem of switches disregarding vacancies in the flow table when deleting expired flow entries. When a table-miss occurs, the corresponding switch must perform additional interactions with an OpenFlow controller to insert new flow entries, but this results in additional processing time and communication overhead. Previous studies aimed at solving this limitation have used sophisticated training sets or modified network architectures. In contrast, we propose a simple flow table management scheme with a least recently used-based caching algorithm to keep flow entries in an OpenFlow switch as long as possible. For this purpose, the switch continually adjusts its cache size according to the vacancy of each flow table and the controller determines the packet forwarding path through the switches by referring to the vacancies. We perform an experimental evaluation of the proposed scheme and demonstrate the performance gains in terms of the number of table-missed packets. We also analyze the effectiveness of the flow entry caching scheme using a mathematical model.

Design and Demonstration of High-Throughput Protocol Oblivious Packet Forwarding to Support Software-Defined Vehicular Networks

Software-defined networking (SDN) is a promising technology that can resolve the challenges faced by vehicular networks. However, the OpenFlow-based SDN implementations can only provide a protocol-dependent data plane. This can restrict the effectiveness of software-defined vehicular networks, since special-purpose protocols that have not been standardized in OpenFlow specifications are used frequently in vehicular networks. To address this issue, this paper studies how to realize a protocol-independent data plane by leveraging the protocol oblivious forwarding (POF). Specifically, we present the design of a software-based POF switch (PVS) that supports runtime protocol customization in principle. We implement PVS in a switch box, and propose a flow table management scheme to ensure high-throughput packet forwarding. The experimental results verify that PVS can achieve line-rate packet forwarding at 10 Gbps when the packets’ size is 512 B, and the proposed flow table management scheme is effective.

Proactive-LRU: An Efficient Flow Table Management Scheme in SDNs

A Software Defined Network (SDN) separates an existing network structure into a control plane and a data plane and centrally manages the network at the control plane. SDN is able to control the network in detail and operate efficiently by managing the traffic by flow unit. However, if the performance of the switches comprising the SDN varies greatly from one to another, it has a problem of delayed handling of traffic due to the bottleneck caused by an overhead concentration at the controller when the message exchanges between the low performance switch and the controller for flow table replacement increase in number. This research paper suggests a flow table management scheme, called Proactive-LRU, to alleviate this problem and compares its performance to that of the existing replacement algorithms such as LRU, LFU, and FIFO. Proactive-LRU scheme is a flow table management scheme that deletes the flow from all switches available within the forwarding path when a flow is deleted and reduces the number of message exchanges and the controller overhead. In order to compare the performance of the suggested scheme against that of the existing algorithms (FIFO, LRU, LFU), video traffic and web browsing traffic have been mixed in certain ratios in an SDN environment for the experiment. The analysis of the results showed that the number of message exchanges as well as the overhead and table overflow of switches in the suggested scheme were significantly reduced.

JumpFlow: Reducing flow table usage in software-defined networks

The forwarding scheme in Software-Defined Networking (SDN) is usually coupled with flow table management. To reduce the redundancy in the flow tables of OpenFlow switches, some recent studies propose forwarding flows using stacked MPLS labels, in which each label in the stack indicates the forwarding decision at one hop of the forwarding route. However, using multiple MPLS labels in each packet introduces significant transmission overhead, especially in networks with large diameters.In this paper, we propose JumpFlow, a forwarding scheme that achieves low and balanced flow table usage in an SDN by properly and reactively placing flow entries on switches. To reduce the transmission overhead, JumpFlow uses the available VLAN identifier (VID) in the packet header to carry routing information. Constrained by the limited space of the VID, a flow’s complete routing information must be divided into several sections and loaded separately at different switches on the flow’s forwarding route. To achieve low and balanced flow table usage, we formulate and solve the reactive flow entry placement problem. We evaluate JumpFlow against the per-hop configuration-based forwarding of OpenFlow for both unicast and multicast scenarios in a real network topology with different traffic patterns. For the unicast scenario with different new flow arrival rates, JumpFlow postpones the time when the first flow rejection occurs, reduces the flow rejection percentage by 37.06%, and reduces the control messages for route configuration by 53.52% on average. For the multicast scenario with a high new multicast group arrival rate, JumpFlow increases the ratio of accepted multicast groups by 83.90%, and reduces the ratio of average control messages for a multicast group configuration by 32.68%.