Inhalt des Dokuments
- © Fraunhofer FOKUS
- Research Area: Measurement and Monitoring
- Projects: OpenLab , Fed4FIRE , and G-Lab 
Multipoint Measurement - Packet Tracking
- © Copyright??
Nowadays more and more networks applications (like multimedia and gaming) demand a stronger quality from the network like small latencies or low packets loss. In order to supervise the fulfillment of such demands, we need not only a good understanding of the network itself but also need to know what happens in it. We have realized a distributed network surveillance solution for detecting paths of data packets through networks. The Multi-Hop Packet-Tracking permits to “track” packets in the network and answer questions like: The routes followed by the packets through the network, the packets delays and losses between nodes, the service chains used by some data traffic, the influence of cross traffic, and the network QoS on the observed network links. We note that this Packet-Tracking solution needs to take care of the network performance and that the packet selection must be in a way so that we observe the same set of packets in the whole network. Multi-Hop Packet Tracking uses a non-intrusive coordinated sampling method [RFC5475] to show the hop-by-hop path and transmission quality of data packets from arbitrary applications. The packet tracking solution can provide means to verify (multipath) routing and overlay experiments and provide input to adaptive network algorithms, security applications and the supervision of service level agreements.
Packet Tracking Visualization
Visualization GUI (Netview). It receives and visualizes measurement data by providing a graphical output of packet tracking paths to facilitate analysis of processed data. For more details see video .
- Lean, efficient packet tracking measurement by employing standards
- Coordinated hash-based sampling [RFC5475]
- Resource control through sampling rate adaptation
- Standardized IPFIX data export [RFC5101]
- Network supervision (management, security, SLA validation, planning)
- Support for experimental network research (experiment verification, routing, P2P) in large scale federated testbeds
- Input for adaptive algorithms (overlay establishment, media streaming, management)
Download the packet tracking software here 
The measurement probe is available on sourceforge.net/projects/impd4e/develop 
A video showing the instantiation and use of the packet tracking system can be found here .
Packet Tracking in Projects
Packet Tracking in the national project G-Lab-Deep 
Packet Tracking allows to “track” single packets in the network and answer enquiries related to routes followed by any selected packet across multiple hops through the network, on topics such as packet delays and losses between nodes, service chains used by some data traffic, cross traffic influence, and network QoS. For this purpose it was used in G-Lab-Deep project.
PDF , Video 
Packet Tracking in the EU FIRE OpenLab project 
Through the resource federation framework Teagle , users can get Infrastructure-as-a-Service distributed across heterogeneous domains, Teagle-based testbeds, Slice Federation Architecture based testbeds, etc. TEAGLE was extended to offer Monitoring-as-a-Service. Packet Tracking is one of the offered monitoring tools.
Packet Tracking Teagle integration was presented as a demo in the OpenLab interim review meeting in Thessaloniki, June 2012. Video 
J. Mueller, Y. Al-Hazmi, M. F. Sadikin, D. Vingarzan, and T. Magedanz. Secure and Efficient Validation of Data Traffic Flows in Fixed and Mobile Networks. Proceeding of the 7th ACM workshop on Performance monitoring and measurement of heterogeneous wireless and wired networks 2012 (PM2HW2N'12), Paphos, Cyprus, Oct 2012. PDF 
Y. Al-Hazmi, J. Mueller, H. Coskun, and T. Magedanz. Network Path and Quality Validation in the Evolved Packet Core. The 12th Würzburg Workshop on IP: ITG Workshop "Visions of Future Generation Networks" (EuroView2012), Würzburg, Germany, July 2012. PDF 
Tanja Zseby, Thomas Zinner, Kurt Tutschku, Yuval Shavitt, Phuoc
Tran-Gia, Christian Schwartz, Albert Rafetseder, Christian Henke,
Multipath Routing Slice Experiments in Federated Testbeds in
Future Internet Assembly (FIA) Book Future Internet: Achievements and Promising Technology. Springer, pages 247-258. May 2011. Link 
Tacio Santos, Christian Henke, Carsten Schmoll, Tanja Zseby. Multi-hop packet tracking for experimental facilities. SIGCOMM '10: Proceedings of the ACM SIGCOMM 2010 conference. ACM, 447–448. September 2010. Link 
Carsten Schmoll, Christian Henke, et al. G-Lab Deep: Cross-Layer Composition and Security for a Flexible Future Internet. In Proceedings of TRIDENTCOM 2010 - 6th International ICST Conference on Testbeds and Research Infrastructures for the Development of Networks & Communities. ICST, May 2010, pages 594-596. ISBN: 978-963-9799-92-9. Link 
Tanja Zseby, Christian Henke, M. Kleis (2010). Packet Tracking in PlanetLab Europe - A Use Case. In Proceedings of TRIDENTCOM 2010 - 6th International ICST Conference on Testbeds and Research Infrastructures for the Development of Networks & Communities. ICST, May 2010, pages 265-274. ISBN: 978-963-9799-92-9. Link 
Christian Henke, Carsten Schmoll, Tanja Zseby. Empirical Evaluation of Hash Functions for PacketID Generation in Sampled Multipoint Measurements. 10th Passive and Active Monitoring Conference 2009, 197-206. Seoul, Korea. Link 
Christian Henke, Carsten Schmoll, Tanja Zseby. Empirical
evaluation of hash functions for multipoint measurements. ACM
SIGCOMM Computer Communication Review (CCR), Volume 38 Issue 3, July
2008, pages 39–50. Link 
Christian Henke, Carsten Schmoll, Tanja Zseby. Evaluation of Header Field Entropy for Hash-Based Packet Selection. 9th Passive and Active Monitoring Conference (PAM) 2008, 82-91. Cleveland, USA. April, 2008.
Next Generation Networks
+49 30 3463 7366
e-mail query 
Faculty IV – Elektrotechnik und Informatik
sec. HFT 7
Tel: +49 30 314 21454
Fax: +49 30 314 26769
d.floroiu [at] tu-berlin.de