Estimating flow distributions from sampled flow statistics
By Nick G. Duffield, Carsten Lund, Mikkel Thorup. IEEE/ACM Transactions on Networking, 2005
314 citations recorded by the source corpus.
Subfields: Artificial Intelligence, Computer Networks and Communications
View Paper: https://www.semanticscholar.org/paper/25d2831316dff79dced24790406cee4dbf09134a
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Nina Taft
- Mark Crovella
- Christophe Diot
- Anukool Lakhina
- Charu C. Aggarwal
- Thanh T. L. Tran
- Divesh Srivastava
- Graham Cormode
In Computer Networks and Communications
- Mark Crovella
- Christophe Diot
- Anukool Lakhina
- Nina Taft
- Graham Cormode
- Charu C. Aggarwal
- Thanh T. L. Tran
- Divesh Srivastava
Citations in the corpus
- Characterization of network-wide anomalies in traffic flows (2004, methodology): “In [6], Duffield et al proposed techniques to infer the population of worm infections from sampled traffic measurements.”
- Structural analysis of network traffic flows (2004, background): “As [8, 10] show, these unsampled flows have a small number of packets, carry very few bytes and so will have negligible impact on our aggregated flow bytecounts.”
- FlowRadar: A Better NetFlow for Data Centers (2016, background): “Fine-grained traffic analysis: Previous research has shown that packet sampling is inadequate for many finegrained monitoring tasks such as understanding flow size distribution and anomaly detection [22, 20, 30].”
- Flow Monitoring Explained: From Packet Capture to Data Analysis With NetFlow and IPFIX (2014, unknown): “Methods to estimate sampled flow statistics have been discussed in [42].”
- Differentially Private Publication of Sparse Data (2011, methodology): “However, we can adjust for this, e.g., by rounding negative values up to the smallest meaningful value, 0.”
- A Survey on Internet Traffic Identification (2009, background): “A variety of sampling strategies have been proposed recently for optimizing the packet selection process (for fl ow accounting) [20][21] and fl ow selection (for statistical analysis of the original traf fi c) [22].”
- Building a better NetFlow (2004, unknown): “All the router operator has to do is to specify the reported number of (cid:13)ow records M desired for each one minute measurement bin.”