A measurement study of correlations of Internet flow characteristics

Lan, Kun-Chan; Heidemann, John

doi:10.1016/j.comnet.2005.02.008

Cited by 148 publications

(119 citation statements)

References 21 publications

Supporting

Mentioning

114

Contrasting

Unclassified

Order By: Relevance

“…As we will show later in Section 3, this parameter has a significant impact on our classification results. In order to identify elephant flows, we use the metric proposed in [45], where a flow is considered as elephant when its size is greater than the mean flow size observed in the trace plus three times the standard deviation. Figure 2 plots the application breakdown according to L7-filter of the traces in our evaluation dataset.…”

Section: Evaluation Datasetsmentioning

confidence: 99%

Analysis of the impact of sampling on NetFlow traffic classification

et al. 2011

View full text Add to dashboard Cite

The traffic classification problem has recently attracted the interest of both network operators and researchers. Several machine learning (ML) methods have been proposed in the literature as a promising solution to this problem. Surprisingly, very few works have studied the traffic classification problem with Sampled NetFlow data. However, Sampled NetFlow is a widely extended monitoring solution among network operators. In this paper we aim to fulfill this gap. First, we analyze the performance of current ML methods with NetFlow by adapting a popular ML-based technique. The results show that, although the adapted method is able to obtain similar accuracy than previous packet-based methods (≈90%), its accuracy degrades drastically in the presence of sampling. In order to reduce this impact, we propose a solution to network operators that is able to operate with Sampled NetFlow data and achieve good accuracy in the presence of sampling.

show abstract

Section: Evaluation Datasetsmentioning

confidence: 99%

Analysis of the impact of sampling on NetFlow traffic classification

et al. 2011

View full text Add to dashboard Cite

show abstract

“…The use of separate state machines for each variant however proves unscalable given the many flavours of TCP congestion control which have since been deployed. In [20], Lan et al analyse flows according to size, duration, rate and burstiness and characterise the observed correlations for heavy-hitters specifically, uncovering evidence of increased application influence on flow rates and burstiness and consequently suggest treating flow size and duration as independent dimensions.…”

Section: Related Workmentioning

confidence: 99%

On rate limitation mechanisms for TCP throughput: A longitudinal analysis

et al. 2017

View full text Add to dashboard Cite

TCP remains the dominant transport protocol for Internet traffic. It is usually considered to have its sending rate covered by a sliding window congestion control mechanism. However, in addition to this normal congestion control, a number of other mechanisms limit TCP throughput. This paper analyzes the extent to which network, host and application settings define flow throughput over time and across autonomous systems. Our study draws on data from a longitudinal study spanning five years of passive traces collected from a single transit link. Mechanisms for this include limiting by application, interference with the TCP window control mechanism and artificial limitations on maximum window sizes by the operating system. This paper uses a large data set to assess the impact of each mechanism. We conclude that more than half of all heavy-hitter inbound traffic remains throttled by constraints beyond network capacity. For this data set, TCP congestion control is no longer the dominant mechanism that moderates throughput.

show abstract

“…The potential flow behavior parameters that are relevant to our research are: 16] has shown that some large flows may also be long in duration, normally presenting a heavy-tail distribution. Duration can therefore be a potential input parameter for our automatic decision approach.…”

Section: List Of Potential Network Parametersmentioning

confidence: 99%

A Statistical Analysis of Network Parameters for the Self-management of Lambda-Connections

Fioreze

Granville

Sadre

et al. 2009

Scalability of Networks and Services

View full text Add to dashboard Cite

Abstract. Network monitoring plays an important role in network management. Through the analysis of network parameters (e.g., flow throughput), managers can observe network behavior and make decisions based on them. The choice of network parameters although should be relevant for each specific objective. In this paper, we focus on the analysis of network parameters that are relevant for our self-management of lambda-connections proposal. This proposal consists of an automatic decision process to offload large IP flows onto lambda-connections. This paper aims at statistically analyzing a list of potential network parameters as relevant estimators for flow volume. The main contribution of this work is the introduction of a statistical methodology to validate that some few network parameters can be considered as good predictors for flow volume. These predictors are therefore of great interest to be used in our automatic decision process.

show abstract

A measurement study of correlations of Internet flow characteristics

Cited by 148 publications

References 21 publications

Analysis of the impact of sampling on NetFlow traffic classification

Analysis of the impact of sampling on NetFlow traffic classification

On rate limitation mechanisms for TCP throughput: A longitudinal analysis

A Statistical Analysis of Network Parameters for the Self-management of Lambda-Connections

Contact Info

Product

Resources

About