Idle - relay

[TODO: Add link to python script]

This setup was run 5 times to obtain enough data to not get biased by one result. We didn’t run the numbers mentioned in the task description because of a problem we faced with increasing the number of nodes.

Further analysis is required to determine if this issue is related to the one observed in .

From 2 nodes to 23~ it works without any problem. Once that number starts to be higher, it starts to happening that some requests are “lost”, and the node reports:

{'message': 'record not found'}

Delays are obtained calculating the time it takes to send the requests to all nodes from Alice (T1 is the timestamp of the response), until it gets accepted (T2 is the signal timestamp of "@{nodespublic_key} accepted your contact request" in Alice).

Between T1 and T2 there is user interaction, so maybe it would make more sense to measure, if it makes sense:

1. Times between a message is sent until it is received by Bob.
2. Time between Bob accepts, and Alice gets the notification.

Bandwidth, Memory and CPU measurements are taken between the moment the friend requests are sent, until the idle waiting time expires (3 minutes). This could be repeated with higher idle times once the issue with number of nodes is addressed.

Friend request delays:

Time from send to receive acceptance of friend requests in 5 repeated scenarios.

Bandwidth usage:

Comparison of received-sent bandwidth from the entire container on 5 different experiments.

Memory usage: