No abstract

No abstract

This article presents the development of an SNMP v3 agent for user modelling in LAN environments. This agent establishes SNMP communications both with the network managers in charge of configuring the modelling process and with the users from whom it collects information contained in the MIBs (Management Information Base) to find a pattern that characterizes their behaviour. This information will be processed and analyzed by a neural network type SOM (Self Organizing Map), which will allow, after the learning process, the detection of anomalies concerning the normal behaviour of the user. Both the parameters to be configured to define the modelling of each user and the results of the agent's supervision are collected in the modelling MIB contained in the proposed agent. In this way, the developed agent provides a unique tool to model all the users of the same LAN network and constitutes a fully integrated system in the SNMP architecture. Finally, a test scenario is presented for the application of the intrusion detection of the proposed agent.

The demand for ultra-low latency requirements is fueled by the growing popularity of time-sensitive applications including virtual, augmented and mixed reality, and industrial IoT. Edge computing is positioned to fulfill such stringent latency requirements. Addressing the increasing demand for time-sensitive applications becomes challenging due to limited resource at the edge. Even though virtual network function (VNF) sharing is known to improve the utilization of the service providers' resources, service requests -including time-sensitive ones-can nevertheless be rejected. This paper proposes PSVS: a Prediction-based Service placement scheme with VNF Sharing at the edge. PSVS utilizes the predicted required resources in a defined lookahead window to minimize the rejection rate of premium services. A safety-margin is empirically-defined and used to add resiliency against prediction errors. Results show more than a 50% reduction in the rejection rate of premium services. Moreover, PSVS is resilient to prediction errors.

No abstract