Davide Adami

Skype-Hunter: A real-time system for the detection and classification of Skype traffic

In the previous years, Skype has gained more and more popularity, since it is seen as the best VoIP software with good quality of sound, ease of use and one that works everywhere and with every OS. Because of its great diffusion, both the operators and the users are, for different reasons, interested in detecting Skype traffic. In this paper we propose a real-time algorithm (named Skype-Hunter) to detect and classify Skype traffic. In more detail, this novel method, by means of both signature-based and statistical procedures, is able to correctly reveal and classify the signaling traffic as well as the data traffic (calls and file transfers). To assess the effectiveness of the algorithm, experimental tests have been performed with several traffic data sets, collected in different network scenarios. Our system outperforms the ‘classical’ statistical traffic classifiers as well as the state-of-the-art ad hoc Skype classifier. Copyright

On virtualization-aware traffic engineering in OpenFlow Data Centers networks

Oversubscription of intra-Data Center network links and high volatility of VM deployments require a flexible and agile control of Data Center network infrastructures, also integrated with computing and storage resources. In this scenario, the Software-Defined Network paradigm and, specifically, the OpenFlow protocol, opens up new opportunities for the design of innovative resource management platforms that enable dynamic and fine-grain control of DC networks through traffic engineering algorithms. This paper investigates the performance of two different sets of cloud-fluent traffic engineering algorithms. Conceived to work during cloud service deployments, the main target of such algorithms is to achieve a better utilization of network resources by exploiting OpenFlow capabilities for traffic-aware deployments of Virtual Machines. The effectiveness of the proposed solutions is evaluated in terms of network link utilization against VM requests acceptance ratio through simulations and experimental tests carried out by using an ad-hoc emulator.

A network control application enabling Software-Defined Quality of Service

The constant growth in IT as well as the increase of users and services requirements are making communication networking more and more complex to manage. This is why there has been the need to look for new approaches. One very promising solution is Software Defined Networking (SDN) that decouples the data and control planes, having one centralized controller for the network. This gives chances to control and manage the network as desired, thus opening many new possibilities. In this work we chose to exploit SDN to enable QoS control, starting from existing tools (Floodlight) and making modifications in order to better exploit their potential. A new network control application for end-to-end QoS provisioning is built on top of Floodlight and validated in an emulated SDN environment.

Class-based traffic recovery with load balancing in software-defined networks

In the last years, Software-Defined Networking (SDN) has emerged as a flexible paradigm driving innovation in future network architectures. Unfortunately, resilience, a key requirement in IP networks, is not an inherent feature of the SDN architecture. Indeed, the automatic reconfigurability of traffic paths in case of links or nodes failures is no more available. This paper deals with the design and validation of an SDN control application for class-based traffic recovery with load balancing. The features of the OpenFlow protocol have been exploited to timely detect link failures as well as to estimate links utilization. The performance of different strategies for class-based traffic recovery has been evaluated in an emulated environment, based on the widely used Mininet tool, highlighting the effectiveness of the proposed solution in enterprise networks.

A New Path Computation Algorithm and Its Implementation in NS2

Originally conceived as a fast forwarding technique, MPLS provides support for traffic engineering and network survivability. Constrained-based path computation is a key building block for traffic engineering in MPLS networks, since it allows to select a path that satisfies assigned QoS requirements. In this paper, we introduce a novel path computation procedure which aims at improving the performance of the well-known Wang-Crowcroft algorithm by means of some heuristics. Moreover, the two algorithms have been developed in NS2 as an extension of OSPF-TE\ns and integrated with RSVP-TE\ns. Finally, the paper shows how the developed software module can be used to satisfy a set of LSP allocation requests with multiple QoS constraints.

G-RDM: A New Bandwidth Constraints Model for DS-TE Networks

MPLS traffic engineering (TE) allows the creation of end-to-end paths across the network with bandwidth reservations. The main drawback of the basic MPLS-TE model is that it operates at an aggregate level and so it is unaware of traffic classes. DiffServ-aware MPLS-TE (DS- TE) refines the MPLS-TE model by allowing bandwidth reservations to be carried out on a per-class basis. The result is the ability to give strict QoS guarantees while optimizing the use of network resources. Bandwidth constraints models play a key role in the DS-TE architecture, since they establish how bandwidth is distributed among different classes. In this paper, we first present a new bandwidth constraints model, called G-RDM. Then, we compare the performance of G-RDM with respect to MAM and RDM in different scenarios by means of an analytical model based on Markov-chains. The results show that G-RDM, joining the best features of MAM and RDM, allows to improve their performance.

On synchronization techniques: performance and impact on time metrics monitoring

The paper presents an experimental comparison of device synchronization strategies to monitor quality of service (QoS) time metrics, such as the one-way delay and the delay variation. The compared strategies are based on global positioning system (GPS), public Internet network time protocol (NTP) servers and ad hoc GPS-based NTP servers with different treatments of the NTP traffic. The presented experimental results allow to quantitatively evaluate the level of performance achievable using each synchronization technique. Finally, considering a satellite test bed, we focus our analysis on two relevant aspects of time metrics monitoring: the impact of the synchronization errors on the metrics observation and the different values that can be measured when a metric is monitored at the application or data link level. Copyright

An SDN orchestrator for resources chaining in cloud data centers

The oversubscription of intra-Data Center links and the high volatility of Virtual Machine deployments require a flexible and agile control of the Data Center network infrastructure, possibly coordinated with computing and storage resources control. In this scenario, the Software-Defined Networking paradigm opens up new opportunities to design orchestrated resources control strategies able to meet the dynamically-changing traffic demands of instantiated Virtual Machines. This paper presents the architectural design of an SDN-based orchestrator for dynamic computing and communication resources chaining, able to assure better than best effort VMs data exchanges across Cloud Data Centers. Moreover, the paper evaluates a number of different chaining strategies able to dynamically compose proper pool of resources based on their variable load. Such strategies are compared against specific management objectives using both simulations and experiments showing their effectiveness in terms of resource usage and service provisioning rate.

Virtual machines migration in a cloud data center scenario: An experimental analysis

Server virtualization enables dynamic workload management in data centers. However, some aspects of virtualization software technologies, like Virtual Machines (VMs) migration or communications between VMs and storage resources, can lead to huge and unbalanced utilization of the intra-data center network. In this paper, we investigate such issues in an experimental testbed, focusing on the measurement of the traffic overhead due to VMs migration in different operating conditions. Our measurement campaigns highlight that performance is strongly affected by several factors, such as VMs placement, VMs active memory, usage of Jumbo frames instead of Ethernet frames. We also analyze and compare the traffic exchanged between hosts on which VMs are placed, when two different storage systems (i.e., NFS and iSCSI) are used.

Design and performance evaluation of service overlay networks topologies

Nowadays, Internet still lacks of adequate support for QoS-sensitive applications, such as VoIP, Videoconference, and Video-on-Demand. In a large extent, this is due to the fact that Internet was originally designed to provide only a best-effort packet delivery service. In recent years, Service Overlay Networks (SONs) have emerged as a profitable way to leverage these issues without changing the underlying infrastructure. In this paper, we address the topology design problem of a SON from a performance point of view. Since the analytical solution of the problem is too computationally complex, we compare the performance of some well-known topologies and we also propose a new traffic demand aware overlay topology. Through extensive simulations, we investigate the performance of each overlay topology in different network scenarios, taking into account overhead and accepted traffic between the overlay nodes.