Howdy, Stranger!

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


HudsonValleyHost (HVH) new VPS with strange specs
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.

HudsonValleyHost (HVH) new VPS with strange specs

Hello folks,

I've got a VPS with HVH lately and it is advertised as "4 vCPU" on an E3 node. Nice!

https://lowendbox.com/blog/hudsonvalleyhost-4gb-openvz-5mo-500gb-storage-openvz-7-50mo-and-40-off-site-wide

The problem is each vCPU is locked to 848Mhz, which is in reality a quarter of 3400Mhz

So they are giving 1 core at 3400Mhz splitted in 4 ?

See the Geekbench here: http://browser.primatelabs.com/geekbench3/8240643 -- Only one core fully effective... hum!

I opened a ticket and their answer it is IntelSpeedStep technology ? Seriously ?

I've always had positive things to say about Ernie when he was the Chief commander of HVH, aka support was more "helpful".

What do you think ?

Thanks!

«1

Comments

  • drama commences

    Thanked by 1Hxxx
  • Run something CPU intensive for a couple minutes and check to see what the core is reporting?

  • Because here at ColoCrossing™, we strive to keep all of our customers dissatisfied and misled. happy and satisfied. In fact, we have proof - just take a look at how we block literally anything with code!

  • FredQcFredQc Member
    edited February 2017

    @WSS said:
    Run something CPU intensive for a couple minutes and check to see what the core is reporting?

    I've done a couple of Sysbench passes and the results are showing that only one core is effective.

    [root@hvh ~]# sysbench --num-threads=1 --test=cpu --cpu-max-prime=50000 run
    sysbench 0.4.12:  multi-threaded system evaluation benchmark
    
    Running the test with following options:
    Number of threads: 1
    
    Doing CPU performance benchmark
    
    Threads started!
    Done.
    
    Maximum prime number checked in CPU test: 50000
    
    
    Test execution summary:
        total time:                          84.0976s
        total number of events:              10000
        total time taken by event execution: 84.0960
        per-request statistics:
             min:                                  8.30ms
             avg:                                  8.41ms
             max:                                 44.68ms
             approx.  95 percentile:               8.65ms
    
    Threads fairness:
        events (avg/stddev):           10000.0000/0.00
        execution time (avg/stddev):   84.0960/0.00
    
    [root@hvh ~]# sysbench --num-threads=4 --test=cpu --cpu-max-prime=50000 run
    sysbench 0.4.12:  multi-threaded system evaluation benchmark
    
    Running the test with following options:
    Number of threads: 4
    
    Doing CPU performance benchmark
    
    Threads started!
    Done.
    
    Maximum prime number checked in CPU test: 50000
    
    
    Test execution summary:
        total time:                          83.9893s
        total number of events:              10000
        total time taken by event execution: 335.8774
        per-request statistics:
             min:                                  8.33ms
             avg:                                 33.59ms
             max:                                 58.82ms
             approx.  95 percentile:              39.60ms
    
    Threads fairness:
        events (avg/stddev):           2500.0000/0.71
        execution time (avg/stddev):   83.9693/0.01
    

    Ah, you mean the CPU speed, well like I said it's locked at 848Mhz

  • @FredQc said:

    @WSS said:
    Run something CPU intensive for a couple minutes and check to see what the core is reporting?

    I've done a couple of Sysbench passes and the results are showing that only one core is effective.

    Then I'd be pissed. I understand fair use, but if they are advertising 4 vCPUs and only allowing full burst of one, welp- they should say so.

    What's /proc/cpuinfo say, anyhow?

  • WSS said: What's /proc/cpuinfo say, anyhow?

    [root@hvh ~]# cat /proc/cpuinfo 
    processor       : 0
    vendor_id       : GenuineIntel
    cpu family      : 6
    model           : 60
    model name      : Intel(R) Xeon(R) CPU E3-1240 v3 @ 3.40GHz
    stepping        : 3
    microcode       : 16
    cpu MHz         : 848.105
    cache size      : 8192 KB
    physical id     : 0
    siblings        : 8
    core id         : 0
    cpu cores       : 4
    apicid          : 0
    initial apicid  : 0
    fpu             : yes
    fpu_exception   : yes
    cpuid level     : 13
    wp              : yes
    flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good xtopology nonstop_tsc aperfmperf cpuid_faulting pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm arat epb pln pts dtherm tpr_shadow vnmi flexpriority ept vpid fsgsbase bmi1 hle avx2 smep bmi2 erms invpcid rtm xsaveopt
    bogomips        : 6784.84
    clflush size    : 64
    cache_alignment : 64
    address sizes   : 39 bits physical, 48 bits virtual
    power management:
    
    processor       : 1
    vendor_id       : GenuineIntel
    cpu family      : 6
    model           : 60
    model name      : Intel(R) Xeon(R) CPU E3-1240 v3 @ 3.40GHz
    stepping        : 3
    microcode       : 16
    cpu MHz         : 848.105
    cache size      : 8192 KB
    physical id     : 0
    siblings        : 8
    core id         : 1
    cpu cores       : 4
    apicid          : 2
    initial apicid  : 2
    fpu             : yes
    fpu_exception   : yes
    cpuid level     : 13
    wp              : yes
    flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good xtopology nonstop_tsc aperfmperf cpuid_faulting pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm arat epb pln pts dtherm tpr_shadow vnmi flexpriority ept vpid fsgsbase bmi1 hle avx2 smep bmi2 erms invpcid rtm xsaveopt
    bogomips        : 6784.84
    clflush size    : 64
    cache_alignment : 64
    address sizes   : 39 bits physical, 48 bits virtual
    power management:
    
    processor       : 2
    vendor_id       : GenuineIntel
    cpu family      : 6
    model           : 60
    model name      : Intel(R) Xeon(R) CPU E3-1240 v3 @ 3.40GHz
    stepping        : 3
    microcode       : 16
    cpu MHz         : 848.105
    cache size      : 8192 KB
    physical id     : 0
    siblings        : 8
    core id         : 2
    cpu cores       : 4
    apicid          : 4
    initial apicid  : 4
    fpu             : yes
    fpu_exception   : yes
    cpuid level     : 13
    wp              : yes
    flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good xtopology nonstop_tsc aperfmperf cpuid_faulting pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm arat epb pln pts dtherm tpr_shadow vnmi flexpriority ept vpid fsgsbase bmi1 hle avx2 smep bmi2 erms invpcid rtm xsaveopt
    bogomips        : 6784.84
    clflush size    : 64
    cache_alignment : 64
    address sizes   : 39 bits physical, 48 bits virtual
    power management:
    
    processor       : 3
    vendor_id       : GenuineIntel
    cpu family      : 6
    model           : 60
    model name      : Intel(R) Xeon(R) CPU E3-1240 v3 @ 3.40GHz
    stepping        : 3
    microcode       : 16
    cpu MHz         : 848.105
    cache size      : 8192 KB
    physical id     : 0
    siblings        : 8
    core id         : 3
    cpu cores       : 4
    apicid          : 6
    initial apicid  : 6
    fpu             : yes
    fpu_exception   : yes
    cpuid level     : 13
    wp              : yes
    flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good xtopology nonstop_tsc aperfmperf cpuid_faulting pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm arat epb pln pts dtherm tpr_shadow vnmi flexpriority ept vpid fsgsbase bmi1 hle avx2 smep bmi2 erms invpcid rtm xsaveopt
    bogomips        : 6784.84
    clflush size    : 64
    cache_alignment : 64
    address sizes   : 39 bits physical, 48 bits virtual
    power management:
    
  • Ask them if your vzctl has a cpulimit of 25. I wouldn't doubt that this is the case.

    Thanked by 1FredQc
  • Looks like Cpu is limited. Not fair.

    Thanked by 1FredQc
  • See the ticket I have with them:

    One reply took about 11 days, so I almost forgot I had a ticket open with them lol

    I saw the reply when I checked my SPAM folder...

  • mikhomikho Member, Host Rep

    from what I can read in this thread you got 4vCPU. they are however capped @ 848Mhz.

    You got what was advertised but it should have mentioned that you only got access to a 1/4 of the core speed.

    My recommendation; cancel and move away. choose a provider that doesn't hide stuff by leaving important information out of the advert.

  • mikho said: from what I can > read in this thread you got 4vCPU. they are however capped @ 848Mhz.

    You got what was advertised but it should have mentioned that you only got access to a > 1/4 of the core speed.

    My recommendation; cancel and move away. choose a provider that doesn't hide stuff by >leaving important information out of the advert.

    Yeah, I have a few vps with Virmach that comes with capped cpus and I do not have a problem with that. What piss me off is that HVH try to hide this or look innocent like no capping is effective on the vps. Come on... tell the truth at least ;)

  • AnthonySmithAnthonySmith Member, Patron Provider

    Further deception, but what do you really expect, if it was something different I really don't know why.

    Not like anything will be done about it round here either the parent company owns this forum.

    Thanked by 1Riz
  • teamaccteamacc Member
    edited February 2017

    @FredQc said:

    I've done a couple of Sysbench passes and the results are showing that only one core is effective.

    Well, duh:

    [root@hvh ~]# sysbench --num-threads=1 --test=cpu --cpu-max-prime=50000 run
    

    Note the --num-threads=1

    Running this same bench on my online.net atom server (C2750) I get this result:

    sysbench 0.4.12:  multi-threaded system evaluation benchmark
    
    Running the test with following options:
    Number of threads: 1
    
    Doing CPU performance benchmark
    
    Threads started!
    Done.
    
    Maximum prime number checked in CPU test: 50000
    
    
    Test execution summary:
        total time:                          176.7660s
        total number of events:              10000
        total time taken by event execution: 176.7623
        per-request statistics:
             min:                                 17.67ms
             avg:                                 17.68ms
             max:                                 17.84ms
             approx.  95 percentile:              17.68ms
    
    Threads fairness:
        events (avg/stddev):           10000.0000/0.00
        execution time (avg/stddev):   176.7623/0.00
    

    So my atom cores, clocked at 2.4ghz are about half as fast as your cores. Sounds about right. (but only if they speed up when under load, which is what support was trying to tell you)

    Now add --num-threads=4 to your bench, and watch the top/htop output. My sysbench time got down to 44s using --num-threads=4, and even down to 22s using --num-threads=8, given that the server I'm on is 8-core.

  • teamacc said: Now add --num-threads=4 to your bench, and watch the top/htop output.

    Look further down I've already done with 4 cores, which gives the same results.

  • And just for shits n giggles I ran the same command on an E3 1231v3, with 80 seconds as output score (1 thread used). Now, if your server really was clocked down to 25% of its original speed, you'd get around 300+ seconds output. You didn't, so your server is not clocked down (under load)

    So yes, HVH is delivering the performance you want, and no, support is not lying to you.

  • @FredQc said:

    teamacc said: Now add --num-threads=4 to your bench, and watch the top/htop output.

    Look further down I've already done with 4 cores, which gives the same results.

    Well, THAT is interesting. However, given that your 1-thread sysbench did produce proper output, your cpu is most likely not downclocked. Why you're getting 1-core scores for a 4-core vps I dont know.

  • FredQcFredQc Member
    edited February 2017

    teamacc said: So yes, HVH is delivering the performance you want, and no, support is not lying to you.

    Well, if you had read in full, I think that you would not come to this conclusion.

    teamacc said: your cpu is most likely not downclocked

    What I'm saying is that I was assigned one core out of four or the equivalent. 3400 / 4 = 850

    There is some throttling in place, but was not mentioned in first place. And they try to hide it.

  • Just read through the ticket. How many levels of staff do these guys have, and do you get a prize if you make it to the top level?

    Thanked by 1ATHK
  • It has always been 1 Core CPU. That's why you can upgrade to more when you order.

    If they are giving you 4 that amounts to one, no problem there.

    Thanked by 1Nick
  • Well, lets tag the guy that's been doing all the LEB commenting under this offer: @Nick

    Thanked by 1FredQc
  • FredQcFredQc Member
    edited February 2017

    teamacc said: do you get a prize if you make it to the top level?

    Yeah, my prize will be like "We deleted your account and get the fuck out" ;-)

  • @FredQc said:

    teamacc said: do you get a prize if you make it to the top level?

    Yeah, my prize will be like "We deleted your account and get the fuck out" ;-)

    Has HVH teminated clients?

  • WSS said: Has HVH teminated clients?

    I don't know as of yet. I was trying to be a little humoristic :P

  • teamacc said: Just read through the ticket. How many levels of staff do these guys have, and do you get a prize if you make it to the top level?

    I can't stand providers who do that =/ where a single problem ends up becoming a huge thread of ticket replies that don't do anything.

  • I still think it would be common courtesy saying that along with 4vCPUs, that you were given 1/4 core each- but that's just me.

  • @WSS said:
    I still think it would be common courtesy saying that along with 4vCPUs, that you were given 1/4 core each- but that's just me.

    Then why would a single-threaded benchmark score the same as a 4-thread bench?

    Might be a problem with their provisioning system, putting all your vCPUs on the same host-core.

  • NickNick Member, Patron Provider

    FredQc said: I don't know as of yet. I was trying to be a little humoristic :P

    I have increased the CPU Units and cpulimit's from the default SolusVM parameters, any better results now @FredQc?

    Thanked by 1FredQc
  • @teamacc said:

    @WSS said:
    I still think it would be common courtesy saying that along with 4vCPUs, that you were given 1/4 core each- but that's just me.

    Then why would a single-threaded benchmark score the same as a 4-thread bench?

    Might be a problem with their provisioning system, putting all your vCPUs on the same host-core.

    Because bullshit metrics, OpenVZ+NoGiveAShit? I already suggested he ask if there is a cpulimit setup on the vzctl on his VPS; I have no idea if it's gone farther than there. If he was locked to 25%, it works differently depending on CentOS 5 and CentOS 6 for figuring that out, so it certainly seems to be throttled, however I don't deal with OVZ enough to tell you why you'd have the exact same meter of use independent of threads.

  • @Nick said:
    I have increased the CPU Units and cpulimit's from the default SolusVM parameters, any better results now @FredQc?

    Ahh...

    cat /proc/cpuinfo cpu MHz : 3392.421

    execution time (avg/stddev): 21.2116/0.00

    You are the real MVP !

    Why does it takes a LET thread to makes all things works?

    Anyway, thank you.

  • AnthonySmithAnthonySmith Member, Patron Provider

    Nick said: I have increased the CPU Units and cpulimit's from the default SolusVM parameters

    SolusVM defaults?

    You mean: 4 cores/400%/1000 units?

    I think you meant to say, the values we put in to solusvm not the solusvm defaults.

Sign In or Register to comment.