Stochastic simulation is an important aid for the design and performance engineering of computer networks. The credibility of simulative results can, however, be seriously affected by human errors (e.g., inconsistencies in the parameter selection, poor initialization of random generators, bugs in the scripts used for post-processing), which become more and more likely and numerous as the dimension of the set of simulated scenarios (simulation campaign) increases. The occurrence of such errors can be limited by using reliable automation tools, i.e. tools which take care of the above mentioned tasks by using state-of-the-art methodologies. This work describes ANSWER (Automated NS-2 Workflow managER), a simulation workflow automation tool for the Network Simulator (ns-2), explicitly designed for facilitating large-scale simulation campaigns, i.e. those involving many factors. Our framework reduces the space for errors when defining scenarios, controls the execution of a large number of scenarios, and reduces the time overhead required for output data analysis.

No abstract

No abstract

In the last ten years, many circuit-switched networks for voice have been replaced with packet-switched ones. Hence, simulating Voice over IP has become of paramount importance in assessing the performance of a network. However, a sound performance analysis should be carried out in conditions which are as close as possible to a real deployment. In this paper we present enhancements to ns2voip [1], a module for simulating realistic VoIP traffic with the ns2 simulator. In detail, we add new features, i.e. a correlated model for packet generation in a two-way conversation and implementing a set of realistic playout buffers to simulate the behavior of the receiver. Our code is available at http://cng1.iet.unipi.it/wiki/index.php/Ns2voip

Due to the trends of centralizing the E/E architecture and new computing-intensive applications, high-performance hardware platforms are currently finding their way into automotive systems. However, the SoCs currently available on the market have significant weaknesses when it comes to providing predictable performance for time-critical applications. The main reason for this is that these platforms are optimized for averagecase performance. This shortcoming represents one major risk in the development of current and future automotive systems. In this paper we describe how high-performance and predictability could (and should) be reconciled in future HW/SW platforms. We believe that this goal can only be reached in a close collaboration between system suppliers, IP providers, semiconductor companies, and OS/hypervisor vendors. Furthermore, academic input will be needed to solve remaining challenges and to further improve initial solutions.