A Tutorial on Systematic Experimental Design
3. End-to-End Validation
The RSpec file 4node-stitching.rspec
is modified from 4node-final.rspec
created in Step II, so all software has been installed to the slivers already. Open two terminals, and login to the client
and server
, respectively.
Run ping
and traceroute
from the client
to the server
xuanliu@client:~$ ping 192.168.20.10 PING 192.168.20.10 (192.168.20.10) 56(84) bytes of data. 64 bytes from 192.168.20.10: icmp_req=1 ttl=61 time=151 ms 64 bytes from 192.168.20.10: icmp_req=2 ttl=61 time=75.8 ms 64 bytes from 192.168.20.10: icmp_req=3 ttl=61 time=75.7 ms 64 bytes from 192.168.20.10: icmp_req=4 ttl=61 time=75.9 ms ^C --- 192.168.20.10 ping statistics --- 4 packets transmitted, 4 received, 0% packet loss, time 3004ms rtt min/avg/max/mdev = 75.733/94.731/151.416/32.728 ms xuanliu@client:~$ traceroute 192.168.20.10 traceroute to 192.168.20.10 (192.168.20.10), 30 hops max, 60 byte packets 1 router-1-lan0 (192.168.10.10) 0.583 ms 0.595 ms 0.561 ms 2 router-2-stitched0 (192.168.1.2) 38.909 ms 38.902 ms 38.865 ms 3 192.168.2.2 (192.168.2.2) 75.354 ms 75.341 ms 75.301 ms 4 192.168.20.10 (192.168.20.10) 75.501 ms 75.473 ms 75.432 ms xuanliu@client:~$
From the traceroute
result, we found that the delay between the client
and router-1
, the delay between router-3
and the server
were very small (about 0.5ms), this is because the end host and the associated router are from the same aggregate. However, the delay between router-1
and router-2
was about 38ms, and this reflected the delay between the aggregate located at Wisconsin and the aggregate located at New York City, and this shows the delay of the stitched link.