The space complexity of approximating the frequency moments
Journal of Computer and System Sciences
Trajectory sampling for direct traffic observation
IEEE/ACM Transactions on Networking (TON)
Statistical Inference
Improving accuracy in end-to-end packet loss measurement
Proceedings of the 2005 conference on Applications, technologies, architectures, and protocols for computer communications
Practical delay monitoring for ISPs
CoNEXT '05 Proceedings of the 2005 ACM conference on Emerging network experiment and technology
Accurate and efficient SLA compliance monitoring
Proceedings of the 2007 conference on Applications, technologies, architectures, and protocols for computer communications
Load shedding in network monitoring applications
ATC'07 2007 USENIX Annual Technical Conference on Proceedings of the USENIX Annual Technical Conference
Every microsecond counts: tracking fine-grain latencies with a lossy difference aggregator
Proceedings of the ACM SIGCOMM 2009 conference on Data communication
An improved analysis of the lossy difference aggregator
ACM SIGCOMM Computer Communication Review
Measurement and analysis of single-hop delay on an IP backbone network
IEEE Journal on Selected Areas in Communications
Sketching the delay: tracking temporally uncorrelated flow-level latencies
Proceedings of the 2011 ACM SIGCOMM conference on Internet measurement conference
| Hi-index | 0.00 |
One-way packet delay is an important network performance metric. Recently, a new data structure called Lossy Difference Aggregator (LDA) has been proposed to estimate this metric more efficiently than with the classical approaches of sending individual packet timestamps or probe traffic. This work presents an independent validation of the LDA algorithm and provides an improved analysis that results in a 20% increase in the number of packet delay samples collected by the algorithm. We also extend the analysis by relating the number of collected samples to the accuracy of the LDA and provide additional insight on how to parametrize it. Finally, we extend the algorithm to overcome some of its practical limitations and validate our analysis using real network traffic.