Tutorial, 18 May, 9:00 - 12:30
Title: Data management in ETOMIC, SONoMA and NMVO infrastructures
Instructors: Bela Hullar, Sandor Laki, Jozsef Steger, Gabor Vattay
Length of the tutorial: Half day (3 hours)
Intended audience: Intermediate
Background knowledge or skills required: knowledge of active network measurement, basic programming skills, basic knowledge of SQL databases
Abstract
To characterize the structure, dynamics and operational state of the Internet it requires distributed measurements. Although in the last decades several systems capable to do this have been created, the easy access of these infrastructures and orchestration of complex measurements is not solved. This tutorial aims to introduce two different architectural approaches for performing network measurements and an easy access archive, called Network Measurement Virtual Observatory, for storing and publishing measurement data.
The first part of this tutorial will provide an overview of the ETOMIC system which is a network traffic measurement platform with high precision GPS-synchronized monitoring nodes. The infrastructure is publicly available to the network research community, supporting advanced experimental techniques by providing high precision hardware equipments and a Central Management System. Researchers can deploy their own active measurement codes to perform experiments on the public Internet. Recently, the functionalities of the original system were significantly extended and new generation measurement nodes were deployed. We aim to provide an overview of the improved management and measurement systems and introduce the new measurement features focusing on high precision end-to-end experiments through a wide range of examples.
In the second part a web service based network measurement approach will be introduced. The system architecture aims to combine the flexibility of mature network measurement infrastructures such as the PlanetLab or ETOMIC systems with the general accessibility and popularity of public services like Web based bandwidth measurement or traceroute servers. To realize these requirements we developed a multi-layer architecture, called SONoMA, which is based on Web Services and the basic principles of SOA, a very popular paradigm in distributed business application development. Our approach opens the door to combine simple services to perform complex network measurements, handles heterogeneous measurement devices, automatically stores the results in a public database. The design of the architecture enables us to protect the network against malicious users as well. Based on the already available services our prototype system offers, we show how easy it is to construct fancy applications relying on on-demand measurement data.
Online sharing of scientific information has accelerated the research activity in various different domains of science, and this paradigm has been strongly effecting the field of network research for the last years. Using the experiences of similar efforts in other areas of science we have built a prototype for the Network Measurement Virtual Observatory. During the design phase of the observatory the goal was to design a schema that enables the incorporation of a wide range of network measurement data and stimulates sharing of data and knowledge among researchers. Our repository contains not only raw measurement data and the corresponding descriptive parameters, but also the results of data analysis. Observatory users are offered two front ends, a portal and a web service interface. The former one provides an easy-to-use graphical web interface, enhanced with schema browsing capabilities, query syntax checker and history keeping facility. Whereas the web services opens the door to the easy development of "online" network data analysis tools for network research. An important feature of the observatory is that users are granted a certain portion of space to make cross analysis of their results along with the data in the observatory possible.
In this tutorial we are going to sketch the basic capabilities of the three systems, so that potential future users can have the initial bump to take advantage of a clever combination of using them together.
Instructors CVs
Bela Hullar received his MSC diploma in Computer Science in 2007 from the Eotvos Lorand University. He has been working in ELTE Communications Network Laboratory since then, as a PhD student. His research interests are in monitoring and measuring computer networks, developing the appropriate measurement infrastructure and its management system.
Sandor Laki received his MSc degree in computer science from the Eotvos Lorand University, Budapest, Hungary, in 2007. Currently he is working towards his PhD in computer science at the Department of Physics of Complex Systems. His primal research interest includes Internet measurement techniques, adaptive protocols and network modelling, especially IP geolocation and traffic classification.
Jozsef Steger graduated as a biophysicist at the Faculty of Science at Eotvos Lorand University in 2001. Now works as an assistant lecturer at the Dept. of Physics of Complex Systems. Carries out his research activities in the study of communication networks, like the Internet. By using the measurement architecture introduced in the article he made queuing delay tomography experiments possible, which enables us to notice congested hotspots within the network. He is an active member of international project works: Moment and Onelab2.
Gabor Vattay is a professor of physics of complex systems at the Faculty of Sciences of Eotvos University. He received his PhD in 1994 and his DSc title in 2003 from mathematical physics of chaotic systems. His interest is in the complex dynamics arising in natural and manmade systems. Since 1998 his interest shifted toward the dynamics of networked systems. He is the founder and director of the Communication Networks Laboratory established at Eotvos University by Ericsson Research since 2000. Leader of the EU supported Internet measurement effort ETOMIC
www.etomic.org