Skip to content

Instantly share code, notes, and snippets.

@jcvenegas
Last active October 8, 2019 12:43
Show Gist options
  • Save jcvenegas/e8c328a505792d57a9d9f0587d42dcfb to your computer and use it in GitHub Desktop.
Save jcvenegas/e8c328a505792d57a9d9f0587d42dcfb to your computer and use it in GitHub Desktop.
Kata overhead

Test based workloads from: kata-containers/tests#1975

Pod size (using limits): cpus: 4 memory: 2Gbi

Notes: 721.423095: Odd this should be at most 4 cpus cpu stats are in percentage so 400 is = 5 cpus Note that even when we give 5 cpus only 4 are fully used in the host, beacuse k8s limits to 4.

workload test_nginx test_fio test_ntttcp test_stress_ng_cpu
cpu_host_average 295 186 26 398
cpu_guest_average 38 26 12 408
cpu_overhead_average 257 159 13 -10
max_cpu_guest 40.860966 45.695260 30.032272 721.423095
max_cpu_host 304.103421 237.174985 54.225418 410.183909
max_cpu_overhead 265.590949 195.746794 26.446801 10.623169
memory_overhead 132089077 17893341751359703818 136639651 119034716
vm_cpus 5 5 5 5
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment