Difficulties in simulating the internet

By Sally Floyd, Vern Paxson. TNET, 2001

851 citations recorded by the source corpus.

Subfields: Computer Networks and Communications

View Paper: https://www.semanticscholar.org/paper/4e4767e68734f35646c9fb6515e6d1dca0262c7f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

Citations in the corpus

An Analysis of the 1999 DARPA/Lincoln Laboratory Evaluation Data for Network Anomaly Detection (2003, background): “Unfortunately, Internet t raffic is very hard to simulate c orrectly [ 28 ].”
Sizing router buffers (2004, result): “Our results are limited to the finite number of different simulations and experiments we can run, and we can’t prove that the results extend to any router in the Internet [21].”
MAWILab: combining diverse anomaly detectors for automated anomaly labeling and performance benchmarking (2010, background): “However, simulating traffic as diverse as it is on the Internet is notoriously difficult [12], especially for anomalous traffic.”
Rapid Generation of Realistic Mobility Models for VANET (2007, methodology): “To reduce such an overhead, MOVE also provides an interface to automatically generate simulation scripts on the ﬂy based on the parameters that the user inputs into MOVE.”
Event labeling combining ensemble detectors and background knowledge (2014, background): “Firstly, they do not reflect the reality [41] and secondly it is extremely difficult to develop a simulator that generates data close to the ground truth [19].”
Entrusting Private Computation and Data to Untrusted Networks (2013, methodology): “Simjong’s network model is based on ns-2 [23], simplified to abstract away the exact topology of the network.”
Performance evaluation and comparison of Westwood+, New Reno, and Vegas TCP congestion control (2004, background): “When a new protocol is proposed, it is necessary to collect a large set of simulation and experimental results in order to assess its validity and the advantages of its deployment in the real Internet [29].”
Power laws and the AS-level internet topology (2003, background): “Paxson and Floyd consider this problem as a major reason why we do not know how to simulate the Internet [21].”
Learning in situ: a randomized experiment in video streaming (2019, background): “Even with representative throughput traces, accurately simulating or emulating the diversity of Internet paths requires more than replaying such traces and is beyond current capabilities [15, 16, 31, 45].”
Outside the Closed World: On Using Machine Learning for Network Intrusion Detection (2010, background): “The point we wish to convey however is that we are working in an area where insight matters much more than just numerical results.”
Learning nonstationary models of normal network traffic for detecting novel attacks (2002, background): “The IDS data includes audit logs and daily dumps of the operating system files on each victim machine, Solaris BSM (system call) logs, and two network traffic sniffer logs, one "inside" between the router and victims, and the other "outside" between the router and the Internet.”