QoS-adaptation by software agents in the presence of defective reservation mechanisms in the Internet (original) (raw)
Related papers
Tunnel Agents for Enhanced QoS in the Internet
Originally, the Internet delivered best-e ort service quality with respect to end-to-end delay. Recently, extensions such as RSVP have been proposed to provide guaranteed real-time services as well. Unfortunately, network resources, such as routers, do not yet fully support RSVP reservation protocols so that guarantees cannot truly be given. In this paper, we suggest to follow the paradigm of open programmable networks for a more complete QoS provisioning. Reservation gaps or tunnels are dynamically closed by means of a software-agent approach that is exibly deployed for an application oriented QoS support. Agents are dynamically located to such tunnels in order to monitor the tunnels, to provide feed-back information in case of QoS violations, and to decide on possible compensating measures to be taken. One such possible measure is redistribution of user speci ed QoS budgets, like end-to-end delay, among the involved RSVP-capable network components. Re-routing support is another feature achievable by exploiting mobility capabilities of the agent approach based on our Java platform.
Programmable agents for flexible QoS management in IP networks
IEEE Journal on Selected Areas in Communications, 2000
Network programmabillity seems to be a promising solution to network management and QoS control. Software mobile-agents technology is boosting the evolution toward application level control of network functionalities. Code may be deployed in the network dynamically and on-demand for the bene t of applications or application classes. Agents support a dynamic distribution of control and management functions across networks, thus increasing exibility and e ciency. We propose to use mobile-agent technology to overcome some of the problems inherent in current Internet technology. We focus our attention to QoS monitoring, being locally signicant in network sub-domains, and realize a QoS management strategy in response to variations of user, customer of application requirements and of the network state. We describe our experience and the results obtained from our test-bed, where software agents are instantiated, executed, migrated and suspended in order to implement exible QoS management in IP networks.
QoS management in programmable networks through mobile agents
Microprocessors and Microsystems, 2001
In delivering multimedia services, quality of service represents a crucial commitment to be satisfied. Very often it has been considered only from a theoretical point of view, leaving any implementation details out of the discussion, mainly for the lack of concrete possibilities to execute its control and management effectively. Recent technological developments in the networking and distributed programming fields are now opening new challenging scenarios towards the negotiation and guarantee of QoS in the delivery of multimedia services through the network. Active or programmable networks are becoming a reality, and the migration of software components among network nodes seems to be the direction pursued by most of network manufactures. Mobile software agents represent a very attractive approach to the distributed control of computer networks and a valid alternative to the implementation of strategies for the management of QoS. In this paper we present our approach to QoS management through mobile agents. The potentiality of this approach is shown through two application examples. The first one focuses on resource reservation through RSVP in an int-serv scenario, while the second one shows how to provide QoS to aggregated traffic flowing through a virtual network.
Programmable Agents for Flexible QoS Management
Network programmabillity seems to be a promising solution to network management and QoS control. Software mobile-agents technology is boosting the evolution toward application level control of network functionalities. Code may be deployed in the network dynamically and on-demand for the bene t of applications or application classes. Agents support a dynamic distribution of control and management functions across networks, thus increasing exibility and e ciency. We propose to use mobile-agent technology to overcome some of the problems inherent in current Internet technology. We focus our attention to QoS monitoring, being locally signicant in network sub-domains, and realize a QoS management strategy in response to variations of user, customer of application requirements and of the network state. We describe our experience and the results obtained from our test-bed, where software agents are instantiated, executed, migrated and suspended in order to implement exible QoS management in IP networks.
An Intelligent Agent Based QoS Provisioning and Network Management System
2007
The DiffServ architecture provides a scalable mechanism for QoS introduction in a TCP/IP network. DiffServ model is based on the aggregation of traffic flows at an ingress (or egress) point of a network and the IP packet marking for different priority flows, according to several classification criteria. Two approaches exist in the DiffServ architecture: the Absolute and the Relative. In Absolute DiffServ, an admission control scheme is used to provide QoS guarantees as absolute bounds of specific QoS parameters. The Relative DiffServ model provides QoS guarantees per service class expressed with reference to guarantees given to other classes defined. Our study presents a QoS Provisioning & Network Management System. This system is based on an extension of the network management architecture followed and implemented within the scope of the European IST Project MANTRIP. MANTRIP network management system supports quality of service configuration and monitoring in IP networks. Our exten...
Closing the Loop for Dynamic IP QoS Provisioning: A Case Study
32nd IEEE Conference on Local Computer Networks (LCN 2007), 2007
This paper investigates an approach to enabling dynamic reactive quality-of-service (QoS) management for IP networks. We use the information collected from monitoring traffic flows combined with middleware for setting network element configurations to differentiate the service given to traffic. We report on experiences of implementing the QoS management system using XORP (eXtensible Open Router Platform) and Click, the open source software router. The performance evaluation in terms of packet loss, dynamic behaviour of QoS configuration and time to re-configure the XORP+Click router is presented. 1 Some terms used in this paper may be registered trademarks of corporations including Tarari, XACCT, and Cisco. Where used, we intend them to be interpreted in the same way as they are understood in common usage in the IP community.
A scheme for time-dependent resource reservation in QoS-enabled IP networks
2000
A number of distributed applications require communication services with Quality of Service (QoS) guarantees. The QoS provisioning issue in the Internet has been addressed by the IETF with the definition of the Integrated Services (IntServ) and Differentiated Services (Diffserv) frameworks. Resource reservation mechanisms on which these models are based are totally timeunaware. Yet, we believe that, in some cases, associating a time interval to network resource reservations could be useful for both users and network providers. In this paper we present a distributed scheme for time-dependent reservations in QoS-enabled IP networks. We also show how the standard signalling protocol RSVP may support this new reservation style, with only a few minor modifications. Finally, we present a first prototype implementation of the major component of the proposed architecture and we provide some hints on future applicability scenarios of the advance reservation paradigm and its impact on related topics such as policing and charging techniques in QoS-enabled IP networks.
A Proposal for End-to-End QoS Provisioning in Software-Defined Networks
International Journal of Electrical and Computer Engineering (IJECE), 2017
This paper describes a framework application for the control plane of a network infrastructure ; the objective is to feature end-user applications with the capability of requesting at any time a customised end-to-end Quality-of-Service profile in the context of dynamic Service-Level-Agreements. Our solution targets current and future real-time applications that require tight QoS parameters, such as a guaranteed end-to-end delay bound. These applications include, but are not limited to, health-care, mobility, education, manufacturing, smart grids, gaming and much more. We discuss the issues related to the previous Integrated Service and the reason why the RSVP protocol for guaranteed QoS did not take off. Then we present a new signaling and resource reservation framework based on the cutting-edge network controller ONOS. Moreover, the presented system foresees the need of considering the edges of the network, where terminal applications are connected to, to be piloted by distinct logically centralised controllers. We discuss a possible inter-domain communication mechanism to achieve the end-to-end QoS guarantee. 1. INTRODUCTION Internet service providers (ISPs) are striving to innovate their network infrastructures at the pace content providers do with their services. Digital contents are consumed by smart phones and sophisticated terminal stations that continuously evolve together with the applications they host. Interestingly enough, the evolution of the Over-The-Top (OTT) services is mainly happening without the aid of network service providers, within the best-effort data traffic channel in the access networks. Recently, a new plethora of applications requiring a RTT delay of around 1ms have been grouped under the hat of tactile-internet applications: a tactile sensor reads information and a connected system reacts with actuators seen by a human within 1 ms [1]. Although we are still far from achieving end-to-end RTT of around 1ms with wireless communications, ISPs need to be ready to re-architect their software control-plane in order to fully exploit the enormous potentials offered by their infrastructures. The goal of this paper is to present the design and a prototype implementation of a control-plane network application for provisioning dynamic end-to-end QoS profiles to end-user applications. The current adoption of distributed control algorithms forces the use of the same signaling protocol (e.g. RSVP, BGP-LS) in all the data-path nodes, not taking into account the resistances inevitably present between device vendors and between administrative domains. For this reason the Service-Level-Agreements (SLAs) between a service provider and its customers or between providers are still mainly static. Moreover, the experience has shown that the scalability issue of the core network in maintaining per-flow state for resource reservation in each node along a path prevented the diffusion of RSVP and integrated services in general. As discussed in [2], per-flow service treatment does not scale in the Internet core; backbone routers must be fast and only an aggregate behaviour is feasible. Instead, it is important to enable such treatment
2001
1 $EVWUDFW AQUILA, a joint European research project funded by EU, defines and implements a new Resource Control Layer (RCL) to enable dynamic end to end QoS provisioning in IP networks for QoS sensitive applications in a scalable and efficient way. The RCL controls the resource usage in an IP backbone that supports different transport services through the 'LIIHUHQWLDWHG 6HUYLFHV (DiffServ) approach. In the DiffServ model the Admission Control (AC) is performed by the elements at the edge of the network. To gain scalability, the AC decision must be based on locally stored information at these elements – mainly resources availability – and keep low the amount of interactions with the rest of the network. To gain efficiency, the availability of resources within the network must be dynamically tracked to some extent. To meet these twofold objective, in the framework of the AQUILA architecture a distributed provisioning architecture is proposed based on the concept of dynamical Reso...