• Corpus ID: 389146

On the Accuracy and Overhead of Cisco Sampled NetFlow

@inproceedings{Choi2005OnTA,
  title={On the Accuracy and Overhead of Cisco Sampled NetFlow},
  author={Baek-Young Choi and Supratik Bhattacharyya},
  year={2005}
}
Traffic measurement and monitoring are an important first step for network management and traffic engineering. With high-speed Internet backbone links, efficient and effective packet sampling are not only desirable, but also increasingly becoming a necessity. The Sampled NetFlow [10] is a Cisco router’s traffic measurement functionality with static packet sampling for high speed links. Since the utility of sampling depends on the accuracy and economy of measurement, it is important to… 

Figures and Tables from this paper

Packet Sampling for Network Monitoring

TLDR
An overview of all known packet sampling techniques, point out their strengths and weaknesses in terms of reliability and accuracy in the estimation in different network traffic parameters, and provide more details about the adaptive packet sampling, as a way to increase this accuracy.

Building a Better Mousetrap

TLDR
FlexSample is presented, a sampling framework that allows network operators to get the best of both worlds: for a fixed sampling budget, FlexSample can capture significantly more small-volume flows for only a small increase in relative error of large traffic flows.

Increasing data center network visibility with cisco NetFlow-Lite

TLDR
The architecture and implementation of NetFlow-Lite is described, and how it integrates with nProbe to provide a scalable and easy-to-adopt solution to provide the network visibility similar to those delivered by NetFlow.

Fast monitoring of traffic subpopulations

TLDR
FlexSample, a traffic monitoring engine that dynamically extracts traffic from subpopulations that operators define using conditions on packet header fields, is presented, finding that it is able to capture significantly more packets from these sub Populations than conventional approaches.

A Framework For Maximizing Traffic Monitoring Utility In Network

TLDR
The empirical results revealed that the traffic monitoring approach is effective and can be used in real world networking applicaitons and improves Quality of Service (QoS) besides compliance with traffic engineering objectives.

MeasuRouting: A Framework for Routing Assisted Traffic Monitoring

TLDR
A theoretical framework for MeasuRouting is presented and synthetic and practical monitoring applications are presented to showcase the utility enhancement achieved with this approach to intradomain routing.

On the Retention Rate of Top Talkers for Sampled NetFlow

TLDR
The results show that sampling NetFlow does affect the retention rates of IP addresses and top talkers in ranking lists, and the higher the sampling rate is, the greater the impact is.

An information-theoretic approach to network monitoring and measurement

TLDR
A network model and an information theoretic framework are developed and the benefits of compressing traces captured at a single monitoring point are determined, and approaches to achieve those benefits are outlined.

Evaluation of Elephant-based Algorithms for Flow Table Reduction under Realistic Traffic Distributions

TLDR
Simple packet sampling performs surprisingly well on realistic traffic, reducing the number of flow entries by a factor up to 400 with the aim to cover 80% of the traffic, which questions the need for more sophisticated algorithms.

References

SHOWING 1-9 OF 9 REFERENCES

Adaptive packet sampling for accurate and scalable flow measurement

TLDR
An adaptive packet sampling technique for flow-level traffic measurement with a stratification approach that provides unbiased estimation of flow size with controllable error bound, in terms of both packet and byte counts for elephant flows, while avoiding excessive oversampling.

NetFlow: information loss or win?

TLDR
This paper asks the question how and how accurately can one infer information from NetFlow, and is interested in TCP connection summaries and accurately aggregated packet and byte counts.

Elementary Sampling Theory

In connection with the notion of probability we described the following situation: Assume we have a population of observations which are related to certain measurable outcomes. This population is

Elementary Survey Sampling

TLDR
The book elucidates design adequately and illuminates Taguchi’s advances, and the extensive set of case studies that cover each topic from the previous chapters will find appeal for virtually any practitioner regardless of specific field.

Statistics: Theory and Methods

A fresh, less formal treatment of classical theoretical probability and statistics is written at an accessible level, with a Bayesian flavor and an emphasis on data and interesting applications. It

Internet Engineering Task Force, IP Flow Information Export Working Group

  • Internet Engineering Task Force, IP Flow Information Export Working Group

Internet Engineering Task Force Packet Sampling Working Group

  • Internet Engineering Task Force Packet Sampling Working Group

The Cooperative Association for Internet Data Analysis). ARTS. http://www.caida.org/tools/utilities/arts

  • The Cooperative Association for Internet Data Analysis). ARTS. http://www.caida.org/tools/utilities/arts

Adaptive Packet Sampling for Total Load Estimation

  • IEEE International Conference on Communications (ICC'03)
  • 2003