TERENA Networking Conference 2000    
         
         
   

User Level Network Performance Monitoring Programme

Cees de Laat and Hans Blom, University of Utrecht, The Netherlands

In the second half of 1999 there was a discrepancy perceived between the perceived performance of the TEN-155 as monitored by DANTE and that observed by some of its users. To help identify possible causes, the University of Utrecht developed a distributed measurement environment which continuously (once per hour) observes throughput and delay behaviour from several edges of the network.

Most network monitoring tools provide information on how much data was transported by core networks. Many tools exist to measure network performance. Most of those tools are not tailored to visualise that what an end user really is interested in: the goodput. Goodput is defined as actual useful data transfer from/to the user application. Packet losses and retransmissions are of no interest for the goodput, although the round trip time, or, more importantly, task completion time, in the applications may increase. Reports from core networks may show high volume data transfers but these numbers are useless if it is not known what part of the data is retransmission.

The network monitoring package named rTPL was developed at the University of Utrecht to visualizing end user Quality of Service aspects from the (research) network. Currently measurements are performed for several applications including the European research network, a distributed compute cluster system scattered over 5 universities in the Netherlands and the user sites of the collaboratory experiment Dynacore (RE 4005). In the Dynacore project the goal was to see if the current backbones can support the network requirements of a collaboratory, in the Distributed ASCI Supercomputer project the goal was to compare the QoS obtained from an ATM based MBS service to the "normal" over provisioned Internet service. In Dynacore we found that the actual available network resources were insufficient to exploit any form of collaboratory, although this situation has improved a lot recently. In the DAS project we can show that the use of a QoS service is very beneficial for the distributed computing application.

A draft report showing the tool and results on the above mentioned work is available from:

http://www.phys.uu.nl/~wwwfi/nettest

Full Paper - Slides (2.30MB)

   
         
   

 

Return to SessionReturn to Index