next up previous
Next: Acknowledgment Up: New Approach for Automated Previous: Agents Architecture

Conclusions and future work

  This paper presented a methodology that enables the creation of powerful dependency models for various use cases, in a--to a considerable extent--automated way. Thus, it solves the worst problems of existing applications of dependency models by overcoming the lack of up-to-date models. Due to the nature of the new approach it is even able to do so in heterogeneous environments. This paper further described an agent based architecture enabling the modeling of large scale scenarios as they are typically found in nowadays' IT-management-world. This is an important advantage, as those scenarios disallow manual model creation simply due to the huge number of managed objects involved.

For future work of the project, we consider to have a closer look at scalability issues, e.g., to determine the number of objects our approach is able to handle in a single domain, esp. taking into account that bandwidth and other resources should be used for management only in a very careful and restricted way. For extreme scenarios the project will investigate how far the use of resources can be reduced, while still being able to generate models of satisfactory quality. Or in other words, can the neural networks be trained better so they are able to cope with much less grained data?

For the part of the neural networks we consider to work on improvements allowing to distinguish between different types of dependencies. A second point is that--additional to the way it is implemented now, where the IT-administrator is not at all involved in the training process of neural networks--a feedback mechanism from the GUI to the neural agents could help to improve the neural networks and thus the modeling results. However, the pre-trained neural network currently used in our prototype already reliably works for various use case.




 
next up previous
Next: Acknowledgment Up: New Approach for Automated Previous: Agents Architecture
Copyright Munich Network Management Team