Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

That's a surprisingly large overhead. I've not measured that large an impact on AMD, particularly for compute heavy.

Did you profile at all? And have you observed if it's not compute-bound? If it's memory or IO bound it can be due to other virtualization overheads, such as memory encryption.





I will try to profile, but how do you suggest going about it, what to measure? Maybe running some synthethics stressing CPU or memory?

The workload is pure memory/CPU, with very little IO so it's 100% compute bound, with much more emphasis on CPU.


There's a lot of I/O hidden in CPU-bound loads related to fetching, prefetching, caching, etc.

The guys post were commenting on has alot of info on this topic lol



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: