Conclusions

From the measurements in section 4 the following, general conclusions can be given:

1.

The for remote control required bandwidth is not obtained during the busiest hors during daytime. However, the workday average show that the mean performances decreases are not so large as may be expected from the daily values.

2.

The hosts with 100 Mbit/s interfaces clearly give better better results than the hosts with 10 Mbit/s interfaces. Hence the importance to upgrade everywhere the LAN's to 100 Mbit/s.

3.

Participation of hosts at the TEN-155 would help very much in the detection of congestion takes places in the NRN's or in the TEN network. Thereby it is not so important that the TEN hosts have a ``fast'' connection with the TEN network, but that they make it possible to ``isolate'' the TEN section from the NRN's, hence making a better analysis of the results possible.