Currently, the network traffic control systems are mainly composed of the Internet core and wired/wireless heterogeneous backbone networks. Recently, these packet-switched systems are experiencing an explosive network traffic growth due to the rapid development of communication technologies. The existing network policies are not sophisticated enough to cope with the continually varying network conditions arising from the tremendous traffic growth. Deep learning, with the recent breakthrough in the machine learning/intelligence area, appears to be a viable approach for the network operators to configure and manage their networks in a more intelligent and autonomous fashion. While deep learning has received a significant research attention in a number of other domains such as computer vision, speech recognition, robotics, and so forth, its applications in network traffic control systems are relatively recent and garnered rather little attention. In this paper, we address this point and indicate the necessity of surveying the scattered works on deep learning applications for various network traffic control aspects. In this vein, we provide an overview of the state-of-the-art deep learning architectures and algorithms relevant to the network traffic control systems. Also, we discuss the deep learning enablers for network systems. In addition, we discuss, in detail, a new use case, i.e., deep learning based intelligent routing. We demonstrate the effectiveness of the deep learning-based routing approach in contrast with the conventional routing strategy. Furthermore, we discuss a number of open research issues, which researchers may find useful in the future.

State-of-the-Art Deep Learning: Evolving Machine Intelligence Toward Tomorrow’s Intelligent Network Traffic Control Systems

The Internet Protocol.

Recent years, Software Defined Routers (SDRs) (programmable routers) have emerged as a viable solution to provide a cost-effective packet processing platform with easy extensibility and programmability Multi-core platforms significantly promote SDRs’ parallel computing capacities, enabling them to adopt artificial intelligent techniques, ie, deep learning, to manage routing paths In this paper, we explore new opportunities in packet processing with deep learning to inexpensively shift the computing needs from rule-based route computation to deep learning based route estimation for high-throughput packet processing Even though deep learning techniques have been extensively exploited in various computing areas, researchers have, to date, not been able to effectively utilize deep learning based route computation for high-speed core networks We envision a supervised deep learning system to construct the routing tables and show how the proposed method can be integrated with programmable routers using both Central Processing Units (CPUs) and Graphics Processing Units (GPUs) We demonstrate how our uniquely characterized input and output traffic patterns can enhance the route computation of the deep learning based SDRs through both analysis and extensive computer simulations In particular, the simulation results demonstrate that our proposal outperforms the benchmark method in terms of delay, throughput, and signaling overhead

Routing or Computing? The Paradigm Shift Towards Intelligent Computer Network Packet Transmission Based on Deep Learning

Network Traffic Matrix (TM) prediction is defined as the problem of estimating future network traffic from the previous and achieved network traffic data. It is widely used in network planning, resource management and network security. Long Short-Term Memory (LSTM) is a specific recurrent neural network (RNN) architecture that is well-suited to learn from experience to classify, process and predict time series with time lags of unknown size. LSTMs have been shown to model temporal sequences and their long-range dependencies more accurately than conventional RNNs. In this paper, we propose a LSTM RNN framework for predicting Traffic Matrix (TM) in large networks. By validating our framework on real-world data from G ´ EANT network, we show that our LSTM models converge quickly and give state of the art TM prediction performance for relatively small sized models.

A Long Short-Term Memory Recurrent Neural Network Framework for Network Traffic Matrix Prediction

This paper presents NeuTM, a framework for network Traffic Matrix (TM) prediction based on Long Short-Term Memory Recurrent Neural Networks (LSTM RNNs). TM prediction is defined as the problem of estimating future network traffic matrix from the previous and achieved network traffic data. It is widely used in network planning, resource management and network security. Long Short-Term Memory (LSTM) is a specific recurrent neural network (RNN) architecture that is well-suited to learn from data and classify or predict time series with time lags of unknown size. LSTMs have been shown to model longrange dependencies more accurately than conventional RNNs. NeuTM is a LSTM RNN-based framework for predicting TM in large networks. By validating our framework on real-world data from GEANT network, we show that our model converges quickly and gives state of the art TM prediction performance.

NeuTM: A neural network-based framework for traffic matrix prediction in SDN

This paper presents a monitoring application for quality of service parameters, called iQoSStats, dealing with Available Transfer Rate and One-Way Delay The tool is running on iOS5-based iPad terminal Due to the lack of Java support, a proxy server is needed to allow the connection to the cloud computing solution based on OpenNetInf It relies on an information-centric communication paradigm and provides large-scale information dissemination The intention was to use this infrastructure to support cross-layer QoS for the Future Internet

An OpenNetInf-based cloud computing solution for cross-layer QoS: Monitoring part using iOS terminals

The dynamics of modern communication networks, especially regarding the user's behavior, determine the necessity of an effective routing protocol that can ensure high performance. In this paper we present the implementation and performance evaluation of a new dynamic multipath routing solution, called Situation Aware Multipath (SAMP), using the OPNET Modeler. The main features of our solution are load balancing and congestion avoidance. To increase the scalability, the network is divided into multipath routing domains. We defined two types of routers: AR (Adaptive Router) and AMR (Adaptive Multipath Router), with different capabilities depending on their location inside the domain. SAMP offers network robustness and reliability by taking advantage of the path diversity. The performance of the proposed solution is evaluated and compared, in terms of throughput and network resource utilization, with three standard routing protocols: OSPF, ECMP and EIGRP. The simulation results demonstrate that the new proposed multipath routing algorithm can assure increased network resource utilization and also a high transmission quality, even in case of congestion.

Performance evaluation of a situation aware multipath routing solution

In this paper we describe the design principles and the preliminary implementation of a situation aware intra-domain routing solution, called Situation Aware Multipath (SAMP) routing algorithm. This offers an effective method to increase throughput and to achieve load balancing by attempting to prevent congestion in computer networks. The path selection is based on statistical cross-layer Quality of Service information provided by a situation-aware network management application. The originality of this approach lies in the practical implementation of a multipath algorithm which uses physical layer status information as a routing metric. The experimental results show a significant increase in the number of packets delivered to the destination in case of congestion, compared to traditional single-path and an ECMP-like routing solution. The global capacity of the network is improved as the statistical information regarding individual link traffic helps SAMP to avoid routing traffic through congested areas. Our approach can be used for QoS-constraint and real-time multimedia applications in wired networks.

Design principles and practical implementation of a Situation Aware Multipath routing algorithm

This paper aims to prove using an OpenFlow-based testbed that a real gearbox-like routing algorithm selection in runtime is feasible. The proper selection and interworking of flow forwarding schemes for Infrastructure and Service Providers may lead to enhanced performances. Limited resources and/or a partial topology are shared by the owner of the network with the service producer, based on a real-time evaluation of the status using the Floyd-Warshall algorithm for single path and the Ford-Fulkerson for multipath routing. At its level, the Service Provider decides the best forwarding scheme involving the Modified Dijkstra’s algorithm and, whenever multipath is needed, the Ford-Fulkerson algorithm. However due to ownership, security issues, and the business models agreed nowadays, only the Infrastructure Providers are actually performing the routing according to own rules (for its users or in case of emergency) and on behalf of Service Providers, according to their rules. Other combinations of algorithms for the existing or new coming types of providers (e.g. for virtual service infrastructure) can be easily tested now in the proposed software-defined networking tool.

OpenFlow-Based Implementation of a Gearbox-Like Routing Algorithm Selection in Runtime

Providing ubiquitous connectivity to the passengers of public transportation vehicles is an important goal of the communication system designers in the context of fast development of the intelligent transportation systems and of Future Internet communication technologies. Efficient usage of transmission resources available in the heterogeneous networking environment, characteristic to today's wireless networks, and effective routing over these networks with minimal involvement of the operator's core networks are some of the requirements for a fast and cost effective deployment. This paper proposes the architecture of an ubiquitous connectivity system for public transportation, the architecture design being considered on three distinct levels: system, functional and platform level. The proposed system architecture specifies a minimal set of entities required to implement envisaged connectivity solution and based on a functional analysis the subsystems and modules are derived. By mapping the functional architecture on the hardware components intended to be used the platform architecture is developed.

Andrei Bogdan Rus

Papers

An OpenNetInf-based cloud computing solution for cross-layer QoS: Monitoring part using iOS terminals

Performance evaluation of a situation aware multipath routing solution

Design principles and practical implementation of a Situation Aware Multipath routing algorithm

OpenFlow-Based Implementation of a Gearbox-Like Routing Algorithm Selection in Runtime

Seamless connectivity platform architecture for public transportation