0

I have two Ubuntu 16.04 Azure VMs created from the same image, in the same region (WestEurope), configured to be the same size (Standard A1, 1 vcpus, 1.75 GB memory) and created at the same time (days of difference).

Inspecting the CPU info they are completly different, one being a Intel and the other AMD:

First VM

>cat /proc/cpuinfo 
processor   : 0
vendor_id   : AuthenticAMD
cpu family  : 16
model       : 8
model name  : AMD Opteron(tm) Processor 4171 HE
stepping    : 1
microcode   : 0xffffffff
cpu MHz     : 2094.724
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 : 5
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 3dnowext 3dnow rep_good nopl cpuid extd_apicid pni cx16 popcnt hypervisor lahf_lm cmp_legacy cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw retpoline rsb_ctxsw vmmcall
bugs        : tlb_mmatch fxsave_leak sysret_ss_attrs null_seg amd_e400 spectre_v1 spectre_v2
bogomips    : 4189.44
TLB size    : 1024 4K pages
clflush size    : 64
cache_alignment : 64
address sizes   : 44 bits physical, 48 bits virtual
power management:

Second VM

>cat /proc/cpuinfo 
processor   : 0
vendor_id   : GenuineIntel
cpu family  : 6
model       : 63
model name  : Intel(R) Xeon(R) CPU E5-2673 v3 @ 2.40GHz
stepping    : 2
microcode   : 0xffffffff
cpu MHz     : 2394.456
cache size  : 30720 KB
physical id : 0
siblings    : 1
core id     : 0
cpu cores   : 1
apicid      : 0
initial apicid  : 0
fpu     : yes
fpu_exception   : yes
cpuid level : 15
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 ss syscall nx pdpe1gb rdtscp lm constant_tsc rep_good nopl xtopology cpuid pni pclmulqdq ssse3 fma cx16 sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand hypervisor lahf_lm abm pti retpoline fsgsbase bmi1 avx2 smep bmi2 erms xsaveopt
bugs        : cpu_meltdown spectre_v1 spectre_v2
bogomips    : 4788.91
clflush size    : 64
cache_alignment : 64
address sizes   : 44 bits physical, 48 bits virtual
power management:

For me is a big problem because one VM does not have AVX instructions support and a program depending on tensorflow fails to run.

How can this be happening?

Is there no guaranty to get the same VM processor for 2 different VMs?

1
  • 1
    If you want to run Tnesofrlow, you should pay for a VM that delivers that.One of the GPU ones - NV, NC, NCv2, NCv3, ND - not sure they actually are more expensive (as your work will finish significantly faster)
    – TomTom
    Apr 28, 2018 at 12:50

1 Answer 1

4

Microsoft don't have a nice matrix to figure it out across all types, but certain compute types are limited to specific processors.

For example:

  • Dv3 are restricted to Intel Xeon E5-2673 v3
  • A8-A11, H, are restricted to Xeon E5-2670
  • G is restricted to Intel Xeon, but unspecified exactly which ones

Sometimes you can find Microsoft blogs, which is about as official as you can get, that can give you more information. In this case that blog states that A-series VMs are undefined in their CPU architecture.

With all this in mind, the closest match you'll find with guaranteed Intel processor is Standard_D2_v3 at around twice the price.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.