No abstract

Abstract:Traces collected at monitored points around the Internet contain representative performance information about the paths their probes traverse. Basic measurement attributes, such as delay and loss, are easy to collect and provide a means to both build and validate empirical performance models. However, the task of analysis and extracting performance conclusions from measurements remains challenging.Ideally, performance modelling aims to find a set of self-contained parameters to describe, summarise, profile and easy display network performance status at a time. This can result in the provision of meaningful information to address applications in fault and performance management, hence providing input to network provisioning, traffic engineering and performance prediction. In this work we present the Weibull Mixture Model, a method to characterise endto-end network delay measurements within a few simple, accurate, representative and handleable parameters using a finite combination of Weibull distributions, with all the aforementioned benefits. The model parameters are related to meaningful delay characteristics, such as average peak and tail behaviour in a daily profile, and can be optimally found using an iterative algorithm known as Expectation Maximisation. Studies on such parameter evolution can reflect current workload status and all possible network events impacting packet dynamics, with further applications in network management.In addition, a self-sufficient procedure to implement the Weibull Mixture Model is presented, along with a set of matching examples to real GPS synchronised measurements taken across the Internet, donated by RIPE NCC.

Abstract-Recent research efforts have resulted in efficient support for IPv6 in Low power Wireless Personal Area Networks (6LoWPAN), with the "IPv6 Routing Protocol for Low power and Lossy Networks" (RPL) being on the forefront as the state of the art routing approach. However, little attention has been paid to IPv6 multicast for networks of constrained devices. The "Multicast Forwarding Using Trickle" (Trickle Multicast) internet draft is one of the most noteworthy efforts, while RPL's specification also attempts to address the area but leaves many questions unanswered. In this paper we expose our concerns about the Trickle Multicast (TM) algorithm, backed up by thorough performance evaluation. We also introduce SMRF, an alternative multicast forwarding mechanism for RPL networks, which addresses TM's drawbacks. Simulation results demonstrate that SMRF achieves significant delay and energy efficiency improvements at the cost of a small increase in packet loss. We have extended the TCP/IP engine of the Contiki embedded Operating System to support both algorithms. Both implementations have been made available to the community.

In wireless sensor deployments, network layer multicast can be used to improve the bandwidth and energy efficiency for a variety of applications, such as service discovery or network management. However, despite efforts to adopt IPv6 in networks of constrained devices, multicast has been somewhat overlooked. The Multicast Forwarding Using Trickle (Trickle Multicast) internet draft is one of the most noteworthy efforts. The specification of the IPv6 Routing Protocol for Low power and Lossy Networks (RPL) also attempts to address the area but leaves many questions unanswered. In this paper we highlight our concerns about both these approaches. Subsequently, we present our alternative mechanism, called Stateless Multicast RPL Forwarding algorithm (SMRF), which addresses the aforementioned drawbacks. Having extended the TCP/IP engine of the Contiki embedded operating system to support both Trickle Multicast (TM) and SMRF, we present an in-depth comparison, backed by simulated evaluation as well as by experiments conducted on a multi-hop hardware testbed. Results demonstrate that SMRF achieves significant delay and energy efficiency improvements at the cost of a small increase in packet loss. The outcome of our hardware experiments show that simulation results were realistic. Lastly, we evaluate both algorithms in terms of code size and memory requirements, highlighting SMRF's low implementation complexity. Both implementations have been made available to the community for adoption.

No abstract