Migration is an important feature for network virtualization because it allows the reallocation of virtual resources over the physical resources. In this paper, we investigate the characteristics of different migration models, according... more
Migration is an important feature for network virtualization because it allows the reallocation of virtual resources over the physical resources. In this paper, we investigate the characteristics of different migration models, according to their virtualization platforms. We show the main advantages and limitations of using the migration mechanisms provided by Xen and OpenFlow platforms. We also propose a migration model for Xen, using data and control plane separation, which outperforms the Xen standard migration. We developed two prototypes, using Xen and OpenFlow, and we performed evaluation experiments to measure the impact of the network migration on traffic forwarding.
Network testbeds strongly rely on virtualization that allows the simultaneous execution of multiple protocol stacks but also increases the management and control tasks. This paper presents a system to control and manage virtual networks... more
Network testbeds strongly rely on virtualization that allows the simultaneous execution of multiple protocol stacks but also increases the management and control tasks. This paper presents a system to control and manage virtual networks based on the Xen platform. The goal of the proposed system is to assist network administrators to perform decision making in this challenging virtualized environment. The system management and control tasks consist of defining virtual networks, turning on, turning off, migrating virtual routers, and monitoring the virtual networks within few mouse clicks thanks to a user-friendly graphical interface. The administrator can also perform high-level decisions, such as redefining the virtual network topology by using the plane-separation and loss-free live migration functionality, or saving energy by shutting down physical routers. Our performance tests assure the system has low response time; for instance, less than 3 minutes to create 4-node virtual networks.
In this paper we propose SLAPv, an efficient control system for service level agreements (SLAs) in virtual network environments. SLAPv fills a gap in the design of SLA enforcers by introducing the following set of control features: (i)... more
In this paper we propose SLAPv, an efficient control system for service level agreements (SLAs) in virtual network environments. SLAPv fills a gap in the design of SLA enforcers by introducing the following set of control features: (i) verification of physical resource usage, (ii) retrieval of real-time profiles of virtual routers, and (iii) enforcement of the adequate usage of resources. Furthermore, SLAPv punishes virtual networks that do not respect their contracts. The control relies on fuzzy logic to enforce resource allocation according to usage profile of routers and the system load. The punishment depends on the exceeding violating value, and on the system load. Results obtained from a developed prototype show the efficiency of the proposed system under different conditions and strategic policies. Tests performed with RIPv2 running on virtual networks show that, for high load scenarios, misbehaving routers are forced to rapidly enter in conformity to their SLAs in less than five seconds.
Managing computer networks is challenging because of the numerous monitoring variables and the difficulty to autonomously configure network parameters. This paper presents the OpenFlow MaNagement Infrastructure (OMNI), which helps the... more
Managing computer networks is challenging because of the numerous monitoring variables and the difficulty to autonomously configure network parameters. This paper presents the OpenFlow MaNagement Infrastructure (OMNI), which helps the administrator to control and manage OpenFlow networks by providing remote management based on a web interface. OMNI provides flow monitoring and dynamic flow configuration through a service-oriented architecture. OMNI also offers an Application Programming Interface (API) for collecting data and configuring the OpenFlow network. We propose a multi-agent system based on OMNI API that reduces packet loss rates. We evaluate both the OMNI management applications and the multi-agent system performance using a testbed. Our results show that the multi-agent system detects and reacts to a packet-loss condition in less than three monitoring intervals.