The internet is an enormous highly structured, multi-owner network. The absence of central control and the heterogeneity of the adopted technologies pose two major challenges in network management. Commercial and administrative factors discourage different organizations from sharing details of their network topology and performance. This severely limits the availability of up to date global information, hindering the effectiveness of network management.
In contrast, an up to date view of the status of network components (nodes and links) is necessary for security and performance management. This proposal focuses on inference of internal node and link metrics through end-to-end measurements, an approach known as Network Tomography (NT). NT does not require administrative access privileges on the monitored network, as internal network components are monitored indirectly, from the network edges. Moreover, NT is agnostic to network heterogeneity.
However, NT hits the snag of the huge dimension and intractability of the solution space. The inherent variability of traffic scenarios and the failure dynamics pose additional challenges to the problem, making it necessary to take account of information obsolescence.
In this project, we aim at designing a novel approach to network monitoring, where samples (monitoring paths) are selected on the basis of their capability to contribute information to reconstruct the network state. The monitoring system makes decisions on which paths should be probed next, on the basis of their probability to contribute new information, through Bayesian reasoning.
We envision the design of an always-on monitoring system, which progressively determines the most suitable paths to probe, taking account of past observations and their obsolescence. The proposed approach to Bayesian Network Tomography allows to guide decisions in a complex space and to continuously monitor the network in order to localize node failures and link congestion.
For the first time, we propose a monitoring module for autonomous systems and multi-domain networks able to combine together different performance metrics of the elements of a network (namely nodes and links) without the need for direct inspections. Our module represents a very useful tool for enhancing network manageability, allowing to increase resilience and to detect malicious attacks in communication networks. These are tasks of crucial importance, as the world grows more and more connected, and as we increasingly rely on online services for many every day fundamental actions.
The goal of this problem is to advance the existing solutions for the development of monitoring system by considering realistic scenarios that can be easily deployed in heterogeneous networks (i.e., networks consisting of different hardware component that require specific configurations for classical monitoring systems).
As we have seen, many have devoted their efforts in the development of monitoring systems for specific networks and topologies. At the same time, several corporate colossuses employ part of their research and development divisions in the design of systems for monitoring network performance.
Existing results that use Network Tomography for realistic monitoring systems in dynamic environment presented strong limitations. With this project, we aim at intelligently combining together techniques of Network Tomography with progressive decision making. For the first time, we pursue the aim of evaluating multiple performance metrics in topology-agnostic, realistic scenarios where inherent network dynamism is considered.
We believe that the advances brought by our envisioned proposal can be employed also in different contexts. For example, Network Tomography techniques were intelligently adapted in [30] to monitor city traffic and localise vehicles without GPS. We strongly believe that the principles of our envisioned project could be adopted in problems coming from different disciplines in Computer Science.
BIBLIOGRAPHY (1)
[1] Y. Vardi. Network tomography: Estimating source-destination traffic in-tensities from link data.Journal of the American statistical association,91(433):365-377, 1996
[2] M. Coates, R. Castro, R. Nowak, M. Gadhiok, R. King, and Y. Tsang. Maxi-mum likelihood network topology identification from edge-based unicast mea-surements.ACM SIGMETRICS Performance Evaluation Review, 30(1):11:20,2002
[3] B. Eriksson, G. Dasarathy, P. Barford, and R. Nowak. Toward the practicaluse of network tomography for internet topology discovery. In2010 ProceedingsIEEE INFOCOM, pages 1:9. IEEE, 2010
[4] Y. Chen, D. Bindel, H. Song, and R. H. Katz. An algebraic approach topractical and scalable overlay network monitoring. InProceedings of the2004 conference on Applications, technologies, architectures, and protocols for computer communications, pages 55:66, 2004
[5] G. Kakkavas, D. Gkatzioura, V. Karyotis, and S. Papavassiliou. A reviewof advanced algebraic approaches enabling network tomography for future network infrastructures.Future Internet, 12(2):20, 2020
[6] N.G. Duffield. Simple network performance tomography. InACM IMC, 2003
[7] N.G. Duffield. Network tomography of binary network performance characteristics.IEEE Trans. on Inf. Theory, 52, 2006
[8] R. R. Kompella, J. Yates, A. Greenberg and A. Snoeren. Detection and localization of network black holes. IEEE INFOCOM, 2007
[9] N.X. Nguyen and P. Thiran. The boolean solution to the congested IP link location problem: Theory and practice. In IEEE INFOCOM, 2007
[10] L. Ma, T. He, A. Swami, D. Towsley, K. K. Leung, and J. Lowe. Node Failure Localization via Network Tomography. In ACM IMC, 2014