Why such a difference in RTT (ping)?

Question

linux

Hello!
I noticed such a feature, when pinging the switch (D-Link DES3524) I have the following picture:

~$ping switch54
PING switch54.local.net (10.0.1.54) 56(84) bytes of data.
64 bytes from switch54.local.net (10.0.1.54): icmp_seq=1 ttl=30 time=1.34 ms
64 bytes from switch54.local.net (10.0.1.54): icmp_seq=2 ttl=30 time=1.36 ms
64 bytes from switch54.local.net (10.0.1.54): icmp_seq=3 ttl=30 time=1.38 ms
64 bytes from switch54.local.net (10.0.1.54): icmp_seq=4 ttl=30 time=1.39 ms
64 bytes from switch54.local.net (10.0.1.54): icmp_seq=5 ttl=30 time=1.39 ms
64 bytes from switch54.local.net (10.0.1.54): icmp_seq=6 ttl=30 time=1.38 ms
^C
--- switch54.local.net ping statistics ---
6 packets transmitted, 6 received, 0% packet loss, time 5003ms
rtt min/avg/max/mdev = 1.349/1.377/1.397/0.017

And when pinging a computer connected to the same switch, this is (regardless of the port and switch):

~$ping comp172
PING ciot172.local.net (10.1.1.172) 56(84) bytes of data.
64 bytes from ciot172.local.net (10.1.1.172): icmp_seq=1 ttl=128 time=0.672 ms
64 bytes from ciot172.local.net (10.1.1.172): icmp_seq=2 ttl=128 time=0.721 ms
64 bytes from ciot172.local.net (10.1.1.172): icmp_seq=3 ttl=128 time=0.694 ms
64 bytes from ciot172.local.net (10.1.1.172): icmp_seq=4 ttl=128 time=0.672 ms
64 bytes from ciot172.local.net (10.1.1.172): icmp_seq=5 ttl=128 time=0.673 ms
64 bytes from ciot172.local.net (10.1.1.172): icmp_seq=6 ttl=128 time=0.683 ms
^C
--- comp172.local.net ping statistics ---
6 packets transmitted, 6 received, 0% packet loss, time 5006ms
rtt min/avg/max/mdev = 0.672/0.685/0.721/0.038 ms

The difference in RTT is twofold!
How can this be explained?

Answer 1

The fact that responding to pings is not the direct responsibility of the switch, and it does this, most likely, with a low priority. But the transfer of packets between computers is in high priority, respectively, is performed quickly.