Truong Thu Huong

ECODANE: A customizable hybrid testbed for green data center networks

In this paper we propose a new testbed architecture that combines hardware network devices with virtual emulation test environment to improve scalability, flexibility and accuracy. The testbed enables to design and experiment new concepts for energy-efficient data center. It is based on OpenFlow, a Software Defined Networking (SDN) technology that facilitates the deployment of energy-aware protocols and mechanisms.

A multi-criteria-based DDoS-attack prevention solution using software defined networking

Software-Defined Networking (SDN) has become a promising network architecture in which network devices are controlled by a SDN Controller. Employing SDN offers an attractive solution for network security. However the attack prediction and Prevention, especially for Distributed Denial of Service (DDoS) attacks is a challenge in SDN environments. This paper, analyzes the characteristics of traffic flows up-streaming to a Vietnamese ISP server, during both states of normal and DDoS attack traffic. Based on the traffic analysis, an SDN-based Attack Prevention Architecture is proposed that is able to capture and analyze incoming flows on-the-fly. A multi-criteria based Prevention mechanism is then designed using both hard-decision thresholds and Fuzzy Inference System to detect DDoS attack. In response to determining the presence of attacks, the designed system is capable of dropping attacks flows, demanding from the control plane.

Data driven hyperparameter optimization of one-class support vector machines for anomaly detection in wireless sensor networks

One-class support vector machines (OCSVM) have been recently applied to detect anomalies in wireless sensor networks (WSNs). Typically, OCSVM is kernelized by radial bais functions (RBF, or Gausian kernel) whereas selecting Gaussian kernel hyperparameter is based upon availability of anomalies, which is rarely applicable in practice. This article investigates the application of OCSVM to detect anomalies in WSNs with data-driven hyperparameter optimization. Specifically, the information of the farthest and the nearest neighbors of each sample is used to construct the objective cost instead of labeling based metrics such as geometric mean accuracy (G-mean) or area under the receiver operating characteristic (AUROC). The efficiency of this method is illustrated over the IBRL dataset whereas the resulting estimated boundary as well as anomaly detection performance are comparable with existing methods.

OpenFlowSIA: An optimized protection scheme for software-defined networks from flooding attacks

In this paper, we propose an optimized protection mechanism (OpenFlowSIA) for Software-Defined Networks from flooding attacks (Distributed Denial-of-Service) based on Support Vector Machine and our proposed algorithm called the Idle-timeout Adjustment (IA). Our methodology not only utilizes SVM advantages in classification such as high accuracy and little processing time, but also applies effectively the IA algorithm and coherent policies to protect network from resource exhaustion caused by flooding attacks, particularly for the SDN controller and OpenFlow switches. Through comprehensive experiments, the OpenFlowSIA scheme illustrates that it can be an innovative solution to secure and save the network resources under flooding attacks in the Software-Defined Networks.

A reliable analyzer for energy-saving approaches in Large Data Center Networks

Nowadays a big effort has been paid to make energy-efficient Data Center Networks (DCNs). There are several proposed solutions to reduce the energy consumption in DCN. However, for researchers, it is difficult to evaluate the energy saving performance of those approaches in a large-size network due to the testbed/emulation environment limitation. In this paper we construct a Performance Evaluation Simulator (called GreenDC Analyzer) that can model a large DCN of thousands of servers. The tool is shown to have a reliability and flexibility in processing and providing accurate and stable performance evaluations with various switch types, and different energy-saving schemes under the same traffic conditions. The tool also allows us to have a deeper investigation of the energy saving performance for the two well-known energy-saving schemes: Power Scaling and Idle Logic. In the bottom line, the Analyzer could be a useful tool for researchers to study pros and cons of different energy-saving approaches in a data center.

A Variable Sampling Interval EWMA Distribution-Free Control Chart for Monitoring Services Quality

Many data in service quality came from a nonnormal or unknown distribution, hence the commonly-used control charts are not suitable. In this paper, new Arcsine Shewhart Sign and Variable Sampling Interval EWMA (Exponentially Weighted Moving Average) distribution-free control charts are proposed. The procedure does not require the assumption of normal data. A Markov chain method is used to obtain optimal designs and evaluate the statistical performance of the proposed charts. Furthermore, practical guidelines and comparisons with the basic Arcsine EWMA Sign control chart are provided. Results show that the proposed chart is considerably more efficient than the basic Arcsine EWMA Sign control chart. The proposed control charts are illustrated by analysing the service quality of the Vancouver City Call Centre.

Real Time Data-Driven Approaches for Credit Card Fraud Detection

Credit card fraud causes many financial losses for customer and also for the organization. For this reason, in the past few years, many studies have been performed using machine learning techniques to detect and block fraudulent transactions. This paper introduces two real time data-driven approaches using optimal anomaly detection techniques for credit card fraud detection. The efficiency of this method is studied over a real data set from European credit card holders. Our experiments show that our approaches achieved a high-level of detection accuracy and a low percentage of false alarm rate. Our approaches will bring many benefits for the organizations and for individual users in terms of cost and time efficiency.

Software defined networking-based one-packet DDoS mitigation architecture

Nowadays, Distributed Denial of Service (DDoS) attacks get the most attention since volumetric attacks saturate companys networks and associated server infrastructure. In fact, DDoS can occur weekly or daily in a network but many organizations have no systems in place to monitor DDoS traffic so as to be aware if their networks are being attacked. Within that context, we propose to develop an architecture that enables a network a capacity of monitoring traffic on the fly and flexibly applying various detection and mitigation methods in order to reduce DDoS impact on the system shortly after it has happened. We also propose a SDN One-packet DDoS Mitigation (SODM) scheme with an Openflow switch functioning as a gateway to protect the inner server infrastructure. We also analyze Internet traffic to understand its common nature during attack and normal time. Knowledge of the traffic characteristics and the way to derive attack indicators are a critical input for the detection mechanism to work. The defense solution performance is evaluated to be able to cope with DDoS in small real time-scale with an acceptable false positive rate of ~ 6%.

Adaptation method for streaming of CBR video over HTTP based on software defined networking

OpenFlow/Software-Defined Networking (SDN) is a new networking paradigm that virtualizes network infrastructure by decoupling the control and data plane logic of traditional network devices. The controller of SDN has the overall look about network topology and hence provides flexibility to network operators to implement its own routing approaches. However, it could not control the way client works. In this paper, we propose a bitrate adaptation algorithm at the client side in real time combined with a SDN based dynamic path selection for HTTP-based video streaming. Experimental results show that the proposed method can provide users with better Quality of Experience (QoE) with lower number of rerouting, higher average quality and smoother video quality than existing methods.