Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!


is this benchmark result so bad for 4GB 4cores?
New on LowEndTalk? Please Register and read our Community Rules.

All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.

is this benchmark result so bad for 4GB 4cores?

nockernocker Member
edited February 2016 in General

i have vps with 4cores and 4 the result of unixbenach is 463 for single core and 1200 for all cores? the result surprises me since the vps is XEN 4core e3-2650 v3 2,3ghz and has 4gb ram..

is the vps so bad?

Comments

  • Terrible.

    Either your provider heavily limits your CPU/IO, or the system is oversold like heck.

  • What does cat /proc/cpuinfo show?

  • @black said:
    What does cat /proc/cpuinfo show?

    this is the result http://pastebin.com/ygzDAjSV

    i tried to paste the result here but my pasted is not written well here

  • Name the provider!

  • @nocker said:

    I've had similar cases in the past, & writing a few support tickets solved some of them (the ones where the cause is not enough 'units' supplied or throttled CPUs, rather than oversold). I guess the advertising looks much better with high # of CPU & most people probably wouldn't even notice if they were heavily throttled or limited

  • i think the result is accurate, i have vps on other provider 1 core 765 ram and has better unixbench result and running faster than my 4gb ram 4 cores..

  • @zevus said:
    I've had similar cases in the past, & writing a few support tickets solved some of them (the ones where the cause is not enough 'units' supplied or throttled CPUs, rather than oversold). I guess the advertising looks much better with high # of CPU & most people probably wouldn't even notice if they were heavily throttled or limited

    i'm considering to find another provider now since the actual performance is terrible too

  • vfusevfuse Member, Host Rep

    Depends on what you're paying for it as well.

  • AnthonySmithAnthonySmith Member, Patron Provider
    edited February 2016

    Really depends, is it HVM or PV, if HVM did you enable PV on HVM drivers, did you verify they are loaded, what does the cpu show up as in the guest, which instructions are loaded vs native, what is the avg .st in top.

    Xen never gets as good UB results as KVM or ovz though, but either way the most likely reason is you are in the same cpu pool as another vps that is hammering the cores while your running your test, or the host has some monitoring in place that reduces your cpu weighting and renices your block device when you start abusing IO which UB really does do.

    The other part of the issue is people buying a vps and expecting dedicated resources and assume even in Xen that applies to IO time.

    Thanked by 1GM2015
  • Run a geekbench that will expose it all performance wise.

    Thanked by 1AnthonySmith
  • @AnthonySmith said:
    Really depends, is it HVM or PV, if HVM did you enable PV on HVM drivers, did you verify they are loaded, what does the cpu show up as in the guest, which instructions are loaded vs native, what is the avg .st in top.

    Xen never gets as good UB results as KVM or ovz though, but either way the most likely reason is you are in the same cpu pool as another vps that is hammering the cores while your running your test, or the host has some monitoring in place that reduces your cpu weighting and renices your block device when you start abusing IO which UB really does do.

    The other part of the issue is people buying a vps and expecting dedicated resources and assume even in Xen that applies to IO time.

    it's XEN PV, unixbench and performance is terrible but io speed around 450mb/s (dd test)

  • bacloudbacloud Member, Patron Provider

    Performance is low. How much do you pay for this VPS?

  • AnthonySmithAnthonySmith Member, Patron Provider

    nocker said: it's XEN PV, unixbench and performance is terrible but io speed around 450mb/s (dd test)

    You are doing sequential writes I guess so probably just hitting a raid cache anyway when doing dd.

    I can only assume your in an over used cpu pool then if they even use cpu pools, ask the host to change your affinity or weight and try again.

Sign In or Register to comment.