Fast accurate computation of large-scale IP traffic matrices from link loads

By Yin Zhang, Matthew Roughan, Nick G. Duffield, Albert G. Greenberg. Measurement and Modeling of Computer Systems, 2003

652 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/0849e21a444d4a3bbea735a788628bea5543f900

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

In Electrical and Electronic Engineering

Citations in the corpus

Sensitivity of PCA for traffic anomaly detection (2007, background): “These issues could potentially be addressed by the technique proposed in [23] for conceptually combining routers according to topology, but we have not evaluated its effectiveness in this context.”
Combining filtering and statistical methods for anomaly detection (2005, background): “However in the last few years many inference based techniques have been developed (such as [21, 22, 18, 17, 19, 6] and many others) that can estimate traffic matrices reasonably well given only per-link data such as SNMP data (that is widely available).”
Providing public intradomain traffic matrices to the research community (2006, neutral): “Providing Public Intradomain Traf.c Matrices to the Research Community Steve Uhlig1, Bruno Quoitin Jean Lepropre2 , Simon Balon2 Computing Science and Engineering Research Unit in Networking Department University of Liege, Belgium University of Louvain-la-Neuve, Belgium {lepropre,balon}@run.monte”
On selfish routing in Internet-like environments (2003, methodology): “In reality,Tt can either be obtained through direct measurements [12] or be estimated bas ed on link loads [44].”
SketchVisor: Robust Network Measurement for Software Packet Processing (2017, background): “This so-called underconstrained problem is commonly found in many matrix interpolation problems [24, 60].”
Diagnosing network-wide traffic anomalies (2004, methodology): “Average quantification esimates for all but one dataset are within 21% of the true size of the anomaly, which is sufficiently accurate for operational settings [31, 32].”
Improving the Scalability of Data Center Networks with Traffic-aware Virtual Machine Placement (2010, methodology): “For D, we consider a hybrid traffic model which combines real traces with the classical Gravity model [27].”
An information-theoretic approach to traffic matrix estimation (2003, methodology): “Also there should be no traffic transiting the network from pe er to peer [28].”
Data streaming algorithms for efficient and accurate estimation of flow size distribution (2004, background): “Recent work [9, 10] show that it is possible to use tomography techniques to infer the traffic matrix from link load and aggregate input/output traffic at each node.”
Structural analysis of network traffic flows (2004, background): “Hence, a thorough understanding of OD .ows is essential for modeling network traf.c, and for addressing a wide variety of problems including traf.c engineering, traf.c matrix estimation, capacity planning, forecasting and anomaly detection.”
GreenTE: Power-aware traffic engineering (2010, methodology): “Once the link load information is collected, the controller computes the network trafﬁc matrix using the tomogravity method, which is lightweight, accurate, and can be done within a few seconds for large ISP networks [8].”