Capturing network traffic with a MAGNeT

Authors:
Jeffrey R. Hay;Wu-chun Feng;Mark K. Gardner
Affiliations:
Computer & Computational Sciences Division, Los Alamos National Laboratory, Los Alamos, NM;Computer & Computational Sciences Division, Los Alamos National Laboratory, Los Alamos, NM;Computer & Computational Sciences Division, Los Alamos National Laboratory, Los Alamos, NM
Venue:
ALS '01 Proceedings of the 5th annual Linux Showcase & Conference - Volume 5
Year:
2001

Citing 4
Cited 2

Wide area traffic: the failure of Poisson modeling

IEEE/ACM Transactions on Networking (TON)
The failure of TCP in high-performance computational grids

Proceedings of the 2000 ACM/IEEE conference on Supercomputing
Bandwidth Monitoring for Network-Aware Applications

HPDC '01 Proceedings of the 10th IEEE International Symposium on High Performance Distributed Computing
The NLAMR network analysis infrastructure

IEEE Communications Magazine

The MAGNeT Toolkit: Design, Implementation and Evaluation

The Journal of Supercomputing
ksniffer: determining the remote client perceived response time from live packet streams

OSDI'04 Proceedings of the 6th conference on Symposium on Opearting Systems Design & Implementation - Volume 6

Quantified Score

Hi-index	0.00

Visualization

Abstract

Current monitoring tools capture network traffic as it appears on the network but are incapable of capturing traffic as it progresses through a running protocol stack. Thus, the current generation of tools cannot record true application-traffic demands and cannot capture protocol-state information at run-time in order to help fine-tune network performance. They also lend no insight into the modulating behavior of protocols (e.g., TCP) that have been shown to impact network performance. In this paper, we introduce MAGNeT -- Monitor for Application-Generated Network Traffic. MAGNeT consists of both Linux kernel modifications and user-application programs. In addition to describing the implementation of MAGNeT, we evaluate its performance and its potential use in applications such as network security, protocol tuning and troubleshooting, and traffic characterization.