| 1 |
> |
| 2 |
> Compare realtime it to measured CPU time. If one realtime second is |
| 3 |
> shorter than a |
| 4 |
> CPU second then you know the host is pausing your VM. There are other ways |
| 5 |
> to |
| 6 |
> check, but this should always work if you can contact an asynchronous time |
| 7 |
> standard. |
| 8 |
> You may need to average the time over tens of seconds or a minute. |
| 9 |
> |
| 10 |
> This method will allow you to figure out that AWS spot instances are |
| 11 |
> oversubscribed ~1.5x. |
| 12 |
> |
| 13 |
> |
| 14 |
Nice. FWIW the guest is running NTP. |
| 15 |
|
| 16 |
So should I run something like: date ; time <some command that runs at |
| 17 |
100%CPU for a minute> ; date ? |
Archives