**6. Conclusion**

In this chapter we have presented an analytical model that represents a multiprocessor traffic monitoring system. This model analyses and quantifies the system performance and it can be useful to improve aspects related to hardware and software design. Even, the model can be extended to more complex cases which have not been treated in the laboratory.

Thus, the major contribution of this chapter is the development of a theoretical model based on a closed queuing network that allows to study the behaviour of a multiprocessor network probe. A series of simplifications and adaptations is proposed for the closed network, in order to fit it better to the real system. We obtain the model's analytic solution and we also propose a recurrent calculation method based on the mean value analysis. The model has been validated comparing theoretical results with experimental measures. In the validation process we have made use of a testing architecture that not only has measured the performance, it has also provided values for some necessary input parameters of the mathematical model. Moreover, the architecture helps to setup tests faster as well as to collect and plot results easier. Ksensor, a real probe, is part of the testing architecture and, therefore, it is directly involved in the validation process. As has been seen in the validation section, Ksensor's throughput is acceptably calculated by the model proposed in this chapter. The conclusions obtained have been satisfactory with regard to the behaviour of the model.

This paper has also come in useful to explain the main aspects of Ksensor, a multithreaded kernel-level probe developed by NQaS research group. It is remarkable that this system introduces performance improving design proposals into traffic analysis systems for passive QoS monitoring.

As a future work, we suggest two main lines: the first one is related to Ksensor and it is about a new hardware-centered approach whose objective is to embed our proposals onto programmable network devices like FPGAs. The second research line aims at completing and adapting the model to the real system in a more accurate way. We are already making progress on new mathematical scenarios which can represent, in detail, aspects such as packet capturing process, congestion avoidance mechanisms between capturing and analysis stages, specific analysis algorithms applied in QoS monitoring and packet filtering.

Finally, it is worth mentioning that the test setup, which has been used to validate the model, will be improved acquiring network hardware at 10 Gbps and installing Ksensor over a server with more than two processors. The model will be tested under these new conditions and we hope to obtain satisfactory results, too.

Thus, further work is necessary to analyse this type of systems with a higher precision, compare their results, in certain conditions, better and prevent us from developing high-cost prototypes.
