Have you considered using the "sar" tool, which periodically collects network, cpu, memory, and other resource data and stores it for later use? I believe most distributions have a 10-minute delay between polls to avoid being to intrusive, and storing too much data.
On my RHEL box I simply installed the "sysstat" RPM which provides the /usr/bin/sar command and waited a few minutes for data to be collected. Then I could review the data like this:
P.S. You can reconfigure the polling frequency by adjusting the crontab file in /etc/cron.d/sysstat
P.P.S. The sar data is kept for 30 days by default. You can review older data by telling sar what file to use with the "-f /var/log/sa/saXY" argument where XY is the two-digit day you're interested in (e.g. 12)
[jcall@nas ~]$ sar -n DEV | grep enp4s0
12:10:01 AM enp4s0 4.86 2.33 1.06 0.25 0.00 0.00 1.11
12:20:01 AM enp4s0 4.70 2.32 1.05 0.26 0.00 0.00 1.07
12:30:01 AM enp4s0 4.73 2.25 1.06 0.27 0.00 0.00 1.01
12:40:01 AM enp4s0 5.32 2.76 1.49 0.39 0.00 0.00 0.97
12:50:01 AM enp4s0 4.57 2.22 1.01 0.25 0.00 0.00 0.90
01:00:01 AM enp4s0 4.71 2.32 1.02 0.27 0.00 0.00 1.08
01:10:01 AM enp4s0 4.60 2.29 1.03 0.28 0.00 0.00 0.97
01:20:01 AM enp4s0 4.63 2.21 1.04 0.26 0.00 0.00 0.94
01:30:01 AM enp4s0 4.68 2.18 1.07 0.25 0.00 0.00 0.97
01:40:01 AM enp4s0 4.70 2.20 1.08 0.25 0.00 0.00 0.99
01:50:01 AM enp4s0 10.28 2.72 1.43 0.35 0.00 0.00 0.88
02:00:01 AM enp4s0 4.69 2.29 1.02 0.27 0.00 0.00 1.05
02:10:01 AM enp4s0 4.80 2.26 1.08 0.25 0.00 0.00 1.03
02:20:01 AM enp4s0 4.77 2.26 1.08 0.26 0.00 0.00 0.96
02:30:01 AM enp4s0 4.96 2.42 1.09 0.33 0.00 0.00 0.97
02:40:01 AM enp4s0 4.99 2.37 1.06 0.32 0.00 0.00 1.08
Average: enp4s0 5.13 2.34 1.10 0.28 0.00 0.00 1.00