Omar Cherkaoui

OpenFlow supporting inter-domain virtual machine migration

Today, Data Center Networks (DCNs) are re-architected in different new architectures in order to alleviate several emergent issues related to server virtualization and new traffic patterns, such as the limitation of bi-section bandwidth and workload migration. However, these new architectures will remain either proprietary or hidden in administrative domains, and interworking protocols will remain in-process of standardization for a time longer than the usually required time to market. Therefore, interworking cloud DCNs to provide the federated clouds is a very challenging issue that seems to be potentially alleviated by a software-defined networking (SDN) approach such as Openflow. In this paper, we propose a network infrastructure as a services (IaaS) software middleware solution based on Openflow in order to abstract the DCN architecture specifities and instantly interconnect DCNs. As a proof of concept we implement an experimental scenario dealing with virtual machine migration. Then, we evaluate the network setup and the migration delay. The use of the IaaS middleware allows automating these operations. OpenFlow solves the problem of interconnecting heterogeneous Data Centers and its implementation offers interesting delay values.

Web services architecture for user control and management of optical Internet networks

One of the primary goals of the CA*net 4 network is to provide end users with the ability, on a peer-to-peer basis, to provision, manage, and control the routing of their own lightpaths across the network without the need to signal or request services from any central network management authority or server. A novel approach to such end-user management and control of lightpaths is described, which uses Web services architecture and grid technology.

Third-generation virtualized architecture for the MVNO context

The third-generation architectures have to support multiple mobile virtual network operators (MVNOs). They have also to host different types of these virtual operators. Virtualizing these architectures will allow the MVNOs to rapidly deploy their equipment. It will separate the management domain between them and the mobile host operator. It will also allow sharing resources and reducing the deployment cost. Motivated by these requirements, we propose some MVNO distributed architectures. First of all, we evaluate the physical and virtual deployment time, then we define the utility function of the equipment for the different types of MVNOs. The utility function evaluates the gain in deployment time for each type of MVNO. This function has to be maximized. Our study demonstrates that the data calls type is the best MVNO candidate for virtualization. This latter consistently yields the best overall utility across an important number of network equipment to be virtualized by varying the time required for software installation and the time spent to determine the physical position of the equipment.

A validation solution for network configuration

The continuous development of the Internet results in an increased number of network services. The network operators and service providers have to deploy various configuration tools to deliver those network services. The different tools used to configure the same equipments can create network configuration inconsistency. Therefore, it is necessary to enhance the configuration tools with validation components. The paper presents a solution for these configuration inconsistencies. Our approach captures the features of the CLIs (command line interfaces), such as the context and parameter dependencies of the commands and the service properties, and translates them into configuration validation rules.

Slices Isolator for a Virtualized Openflow Node

Virtualization is emerging as a key mechanism of scaling the infrastructure and facilitating service deployment for future Internet. Network virtualization provides the ability to slice larger, underutilized physical nodes into smaller, virtual ones. To ensure deterministic performance, guaranteed QoS, and independent behavior, virtual nodes have to be properly isolated. However, a high isolation can have an important impact on the node performance and configuration flexibility. This paper investigates the problem of isolation between the slices of a virtualized switch. Our approach consists on adapting isolation to the desired performance and flexibility while ensuring fairness between the slice. Based on that approach, we developed a model that provide the choice between several levels of isolation. Each level corresponds to one or more resources. We implemented this model for a virtualized Open Flow 1.1 switch. We propose a utility-based evaluation balancing performance and flexibility that provide the adapted isolation level to the required demand. This model gives preference to the user to have isolation with a higher flexibility or lower overhead. Our analysis also shows that the overhead and the resource consumption of the isolation is reduced while remaining flexible and dynamic.

On the Flexibility of MPLS Applications over an OpenFlow-Enabled Network

In todays networks, a node usually has a static role that cannot be easily changed without an expensive upgrade. In MPLS architecture, despite the flexible forwarding data plane not tied to a single forwarding technology, each MPLS node has to be dedicated for a specific role depending on its position in the edge or the core of the MPLS network domain. This paper proposes an approach to address the flexibility of an MPLS node to play multiple roles for different MPLS domains built on top of an underlying OpenFlow-enabled physical network. The pipelined approach through tables introduced in the version 1.1 and later of the OpenFlow specification allows the change of the packet processing behavior by just updating the memory structures -- such as the TCAM and hash tables. It exploits the power of the OpenFlow rules-based paradigm to demonstrate the high level programmability to achieve the deossification of an MPLS infrastructure. In order to validate our proposal, we have implemented our approach over a 100Gbps switch box built on network processors and tested it with three applications to evaluate its flexibility. The results show that the local software to hardware update of a Label Switched Path (LSP) can be made in 2.2ms, in average, and the deployment of a Label Switched Router (LSR) application with 400 labels takes only 392.5ms.

High availability in IMS virtualized network

Virtualization, in telecommunication, has been given great interest over the 20 past years. Migrating operating system instances between physical hosts is one of the most important features of this technology. It allows network administrators to resolve serious problem such us hardware failure, lack in server capacity, while facilitates the management and the upgrade of their systems. With the strong use of Internet Protocol Multimedia Subsystem (IMS) by most of third Generation (3G) network operators as a common infrastructure, live migration may be useful to resolve the overload problem that affects Call Session Control Function (CSCF) components. Our objective is to propose a High availability solution in order to avoid performance degradation. Thus, we aim to proof that when introducing the migration technique within the IMS architecture, we resolve the overload problem and guarantee Quality of Service (QoS) for IMS client. Based on previous work on high availability using virtualization, we describe, in this paper, our methodology and implementation of an experimental scenario using live migration. Then, we evaluate the registration and the session setup delays. The use of migration reduces considerably the delays by 40% for registration and by 38% for session setup. So, we obtain acceptable values compared to the thresholds.

Self-configuration of network devices with configuration logic

Autonomic networking is an emerging approach to the management of computer networks that aims at developing self-governed devices. Among the main issues of autonomic systems is the question of self-configuration. In this paper, we describe a method for discovering and self-generating the configuration of a network device in order to dynamically push a new service into a network. On each configuration, several rules representing the semantics of the services are expressed in a logical formalism called Configuration Logic. From these rules, we show how to use traditional satisfiability methods to automatically generate or modify the configuration of a device with respect to the configuration of its neighbours. We illustrate our case with an example of a switch that automatically discovers its VLAN configuration when connected to an existing network. The results presented here have been implemented into the configuration management tool ValidMaker.

Automated Validation of Service Configuration on Network Devices

Due to the significant development of network services in the past few years, their validation has become increasingly difficult. The advent of novel approaches to the issue of validation is therefore vital for keeping services manageable, safe, and reliable. We present a model for the validation of service configurations on network devices. A service configuration is modelled by a tree structure, and its properties are described by validation rules expressed in terms of these tree elements. By using an existing logical formalism called TQL, we have succeeded in expressing complex dependencies between parameters, and in automatically checking these dependencies against real-world network descriptions in feasible time.

Configuration logic: a multi-site modal logic

We introduce a logical formalism for describing properties of configurations of computing systems. This logic of trees allows quantification on node labels, which are modalities containing variables. We explain the motivation behind our formalism and give both a classical semantics and a new equivalent one based on partial functions on variables.