high steal time on amd epyc-Rome

[Sascha72036]

Hello everyone,

I use the AMD EPYC 7402P to virtualize with Proxmox.
The problem is that from around 45% host-CPU usage the steal time on the VMs increases to more than 5%.

Every VM is started with the CPU flag "host."

root@VM24:~# cat /proc/cpuinfo
processor : 0
vendor_id : AuthenticAMD
cpu family : 23
model : 49
model name : AMD EPYC 7402P 24-Core Processor
stepping : 0
microcode : 0x1000065
cpu MHz : 2799.996
cache size : 512 KB
physical id : 0
siblings : 1
core id : 0
cpu cores : 1
apicid : 0
initial apicid : 0
fpu : yes
fpu_exception : yes
cpuid level : 16
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm rep_good nopl cpuid extd_apicid tsc_known_freq pni pclmulqdq ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm cmp_legacy svm cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw perfctr_core ssbd ibrs ibpb stibp vmmcall fsgsbase tsc_adjust bmi1 avx2 smep bmi2 rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsaves clzero xsaveerptr arat npt nrip_save umip rdpid arch_capabilities
bugs : sysret_ss_attrs spectre_v1 spectre_v2 spec_store_bypass
bogomips : 5599.99
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 40 bits physical, 48 bits virtual
power management:

 

[Hannes Laimer]

Hey,

if the demand for computation increases, then also the time processes have to wait for CPU time will increase. Either
1. the VMs do not have enough CPU resources assigned, or
2. the physical server is over-sold and the virtual machines are competing for resources.

Since the overall usage is at only ~50%, it might help to increase the assigned CPU resources to the VMs.