The deployment of efficient measurement solutions to assist network management tasks without interfering with normal network operation assumes a prominent role in today's high-speed networks attending to the huge amounts of traffic involved. From a myriad of proposals for traffic measurement, sampling techniques are particularly relevant contributing effectively for this purpose as only a subset of the overall traffic volume is handled for processing, preserving ideally the correct estimation of network statistical behavior. In this context, this paper proposes MuST-a multiadaptive sampling technique based on linear prediction, aiming at reducing significantly the measurement overhead and still assuring that traffic samples reflect the statistical characteristics of the global network traffic under analysis. Conversely to current sampling techniques, MuST is a multi and self-adaptive technique as both the sample size and interval between samples are self-adjustable parameters according to the ongoing network activity and the accuracy of prediction achieved. The tests carried out demonstrate that the proposed sampling technique is able to achieve accurate network estimations with reduced overhead, using

Distributed service-oriented traffic control mechanisms, operating with minimum impact on network performance, assume a crucial role as regards controlling services quality and network resources transparent and efficiently. In this way, we describe and specify a lightweight distributed admission control (AC) model which provides an uniform solution for managing QoS and SLSs in multiclass and multidomain environments. Taking advantage of the consensual need of on-line service monitoring and traffic control at the network edges, AC decisions are driven by feedback from systematic edge-to-edge measurements of relevant QoS parameters for each service type and SLS utilization. This allows self-adaptive service and resource management, while abstracting from network core complexity and heterogeneity. In this paper, introducing an expressive notation, we specify the high-level entities for multiservice provisioning in a domain and formalize service-dependent AC equations to assure both intra and interdomain model operation. A proof-of-concept of the AC criteria effectiveness in satisfying each service class commitments while achieving high network utilization is provided through simulation.

Abstract-The trend toward the integration of current and emerging applications and services in the Internet has launched new challenges regarding service deployment and management. Within service management, admission control (AC) has been recognized as a convenient mechanism to keep services under controlled load and assure the required QoS levels, bringing consistency to the services offered.In this context, this article discusses the role of AC in multiservice IP networks and surveys current and representative AC approaches. We address and compare the architectural principles of these AC approaches and their main features, virtues and limitations that impact on the quality control of network services. We identify important design aspects that contribute to the successful deployment of flexible and scalable AC solutions in multiservice networks.

This article proposes a distributed admission control (AC) model based on on-line monitoring to manage the quality of Internet multimedia services and Service Level Specifications (SLSs). The AC strategy covers intra and inter-domain operation, controls quality-of-service (QoS) without adding significant complexity to the network control plane and involves only edge nodes. While ingress nodes perform implicit or explicit AC resorting to service-oriented rules for SLS and QoS parameters control, egress nodes collect service metrics providing them as inputs for AC. The end-to-end approach is viewed as a cumulative and repetitive process of AC and available service computation. We evaluate the AC criterion as regards its ability to ensure service commitments while achieving high network utilization. The results show that the proposed model provides a good compromise between simplicity, service guarantee levels and network usage, even for services with strict QoS requirements. Crucial aspects of the model interrelated areas and implementation key points are also discussed and evaluated.

Achieving an admission control strategy for CoS networks covering both intra-domain and end-to-end operation is still an open issue. This paper discusses how AC can be carried out without adding significant complexity to the network control plane and proposes a distributed service-oriented AC model for these networks. The model only involves the network edge nodes leaving the network core unchanged. Ingress nodes perform implicit or explicit service-dependent AC based on both QoS and SLSs utilization metrics, obtained through edge-to-edge online monitoring performed at egress nodes. From an end-to-end perspective, the flow request is used both for AC and available service computation. Relevant aspects of the model interrelated areas and implementation key points are also discussed.