TY - JOUR
T1 - Server-based non-intrusive measurement infrastructure for enterprise networks
AU - Dong, Yingfei
AU - Hou, Yiwei Thomas
AU - Zhang, Zhi Li
AU - Taniguchi, Tomohiko
N1 - Copyright:
Copyright 2018 Elsevier B.V., All rights reserved.
PY - 1999/8
Y1 - 1999/8
N2 - In this paper we propose a server-based measurement approach that can be deployed in replicated servers within a wide-area enterprise network to provide distributed service to a large number of clients across the Internet. In our approach, each server performs traffic measurement and exchanges the collected metrics with peer servers. This server-based measurement approach consists of three steps: non-intrusive data collection, data analysis and performance metrics generation, and exchange of performance metrics among peer servers. As part of an experimental system we have built, we describe a performance metrics generation tool called Woodpecker, which is designed based on a non-intrusive passive packet capturing mechanism. Experimental results obtained using Woodpecker demonstrate that it is indeed feasible to employ non-intrusive measurement tools to generate the majority of desired performance metrics for dynamic server selection.
AB - In this paper we propose a server-based measurement approach that can be deployed in replicated servers within a wide-area enterprise network to provide distributed service to a large number of clients across the Internet. In our approach, each server performs traffic measurement and exchanges the collected metrics with peer servers. This server-based measurement approach consists of three steps: non-intrusive data collection, data analysis and performance metrics generation, and exchange of performance metrics among peer servers. As part of an experimental system we have built, we describe a performance metrics generation tool called Woodpecker, which is designed based on a non-intrusive passive packet capturing mechanism. Experimental results obtained using Woodpecker demonstrate that it is indeed feasible to employ non-intrusive measurement tools to generate the majority of desired performance metrics for dynamic server selection.
UR - http://www.scopus.com/inward/record.url?scp=4243283743&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=4243283743&partnerID=8YFLogxK
U2 - 10.1016/S0166-5316(99)00023-1
DO - 10.1016/S0166-5316(99)00023-1
M3 - Article
AN - SCOPUS:4243283743
SN - 0166-5316
VL - 36-37
SP - 233
EP - 247
JO - Performance Evaluation
JF - Performance Evaluation
ER -