Abstract. The use of advanced applications has caused end-users to show a new interest in accessing network measurement and status data across multiple domains, in order to detect occasional network problems. Diversity on network technologies, technical solutions and network managers add to the complexity of providing such information, which, when available, includes technical details that are meaningful only to specialists. This paper presents MENTOR, a tool for traffic monitoring where applications’ information are used to provide network performance recommendations in high level language, helping users on network usage. 1. Introduction In recent years we have witnessed a rapid growth of audiovisual application running over the Internet, a cheap alternative of interactive voice and video communication that relies on the adoption of a new generation of powerful and interactive applications. These new kinds of applications require from administrators a quick network problem detection and performance analysis. However, the current monitoring approaches don’t meet these requirements since they concentrate on the administrators the task of interpreting each application performance problem related to the network. In order to overcome these limitations, the monitoring infrastructure should help the administrators diagnose network problems and advise users about network conditions. Current traffic monitoring solutions are far from being scalable and giving the right and meaningful network performance information to network applications and users. In fact, when end-users look for network performance, they are actually concerned about the performance of their applications. However, the monitoring visualization tools usually show performance data without considering any particular user and application characteristics. Network administrators need to count on tools that anticipate the user’s needs, presenting the network performance information in a more customized and productive way. Thus, a significant amount of work still has to be done for the measurement data and interpretation.
[1]
Christina Hattingh,et al.
End-to-End QoS Network Design: Quality of Service in LANs, WANs, and VPNs
,
2004
.
[2]
D. Martin Swany,et al.
PerfSONAR: A Service Oriented Architecture for Multi-domain Network Monitoring
,
2005,
ICSOC.
[3]
Leobino Sampaio,et al.
FLAVOR : A dynamic and open framework for the development of network measurement access and visualization tools
,
2007
.
[4]
Robin Burke,et al.
Knowledge-based recommender systems
,
2000
.
[5]
Tristan Henderson,et al.
Networked games: a QoS-sensitive application for QoS-insensitive users?
,
2003,
RIPQoS '03.
[6]
Stuart E. Middleton,et al.
Capturing knowledge of user preferences: ontologies in recommender systems
,
2001,
K-CAP '01.
[7]
Gediminas Adomavicius,et al.
Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions
,
2005,
IEEE Transactions on Knowledge and Data Engineering.
[8]
José Augusto Suruagy Monteiro,et al.
Implementing and Deploying Network Monitoring Service Oriented Architectures: Brazilian National Education and Research Network Measurement Experiments
,
2007,
2007 Latin American Network Operations and Management Symposium.