F72s_v2 by Microsoft Azure

F72s_v2 is a FSv2 family [Premium Storage capable] [Intel processor] 72 vCPUs server offered by Microsoft Azure with 72 vCPUs, 144 GiB of memory and 590 GB of storage. The pricing starts at 0.3996 USD per hour.
72 vCPU
144 GiB Memory
590 GB Storage
Spare SCore
67,087
(All-cores)
1,455
(Single-core)

Specifications

Server Metadata

Vendor ID
azure
Server ID
Standard_F72s_v2
Name
F72s_v2
API Reference
Standard_F72s_v2
Description
FSv2 family [Premium Storage capable] [Intel processor] 72 vCPUs
Family
FSv2
Hw Virt
Average Time To Start
27
Status
planned-for-retirement
Observed At
2026-10-25T08:35:08.074067

Availability

REGION / IDSPOTONDEMAND
West US 2 (US) / westus2
0.5627 USD/h 3.045 USD/h
West US 3 (US) / westus3
0.5627 USD/h 3.045 USD/h
East US (US) / eastus
0.6699 USD/h 3.045 USD/h
East US 2 (US) / eastus2
1.211 USD/h 3.045 USD/h
Central India (IN) / centralindia
0.5655 USD/h 3.06 USD/h
North Central US (US) / northcentralus
0.5655 USD/h 3.06 USD/h
West India (IN) / westindia
0.5655 USD/h 3.06 USD/h
South Africa West (ZA) / southafricawest
0.5673 USD/h 3.06 USD/h
Jio India Central (IN) / jioindiacentral
- 3.06 USD/h
Jio India West (IN) / jioindiawest
- 3.06 USD/h
Indonesia Central (ID) / indonesiacentral
0.5867 USD/h 3.175 USD/h
Malaysia West (MY) / malaysiawest
0.5867 USD/h 3.175 USD/h
Canada East (CA) / canadaeast
0.6163 USD/h 3.335 USD/h
Canada Central (CA) / canadacentral
1.1159 USD/h 3.335 USD/h
Mexico Central (ZA) / mexicocentral
0.6191 USD/h 3.35 USD/h
South India (IN) / southindia
0.622 USD/h 3.366 USD/h
Korea Central (KR) / koreacentral
0.6387 USD/h 3.456 USD/h
Spain Central (ES) / spaincentral
0.6387 USD/h 3.456 USD/h
North Europe (IE) / northeurope
0.7686 USD/h 3.456 USD/h
Korea South (KR) / koreasouth
0.639 USD/h 3.458 USD/h
Austria East (AT) / austriaeast
0.6453 USD/h 3.492 USD/h
Belgium Central (BE) / belgiumcentral
0.6453 USD/h 3.492 USD/h
Germany West Central (DE) / germanywestcentral
0.6453 USD/h 3.492 USD/h
Poland Central (PL) / polandcentral
0.6453 USD/h 3.492 USD/h
Sweden Central (SE) / swedencentral
0.6453 USD/h 3.492 USD/h
West Europe (NL) / westeurope
0.6453 USD/h 3.492 USD/h
Italy North (IT) / italynorth
0.6474 USD/h 3.492 USD/h
Southeast Asia (SG) / southeastasia
0.652 USD/h 3.528 USD/h
UK South (GB) / uksouth
0.669 USD/h 3.636 USD/h
France Central (FR) / francecentral
0.6719 USD/h 3.636 USD/h
Israel Central (IL) / israelcentral
0.6719 USD/h 3.636 USD/h
UK West (GB) / ukwest
0.6729 USD/h 3.641 USD/h
Central US (US) / centralus
0.6606 USD/h 3.672 USD/h
South Central US (US) / southcentralus
0.6786 USD/h 3.672 USD/h
UAE North (AE) / uaenorth
0.6786 USD/h 3.672 USD/h
West Central US (US) / westcentralus
0.6786 USD/h 3.672 USD/h
Qatar Central (QA) / qatarcentral
0.7026 USD/h 3.802 USD/h
West US (US) / westus
0.7052 USD/h 3.816 USD/h
Switzerland North (CH) / switzerlandnorth
0.7099 USD/h 3.8412 USD/h
Norway East (NO) / norwayeast
0.7106 USD/h 3.845 USD/h
Japan East (JP) / japaneast
0.7118 USD/h 3.852 USD/h
East Asia (HK) / eastasia
0.7185 USD/h 3.888 USD/h
Australia Central 2 (AU) / australiacentral2
0.3996 USD/h 3.996 USD/h
Australia Central (AU) / australiacentral
0.7385 USD/h 3.996 USD/h
Australia East (AU) / australiaeast
0.7385 USD/h 3.996 USD/h
Australia Southeast (AU) / australiasoutheast
0.7409 USD/h 4.009 USD/h
South Africa North (ZA) / southafricanorth
0.7577 USD/h 4.1 USD/h
New Zealand North (NZ) / newzealandnorth
0.7758 USD/h 4.198 USD/h
Japan West (JP) / japanwest
0.7829 USD/h 4.223 USD/h
Chile Central (CL) / chilecentral
0.788 USD/h 4.264 USD/h
Germany North (DE) / germanynorth
0.839 USD/h 4.54 USD/h
UAE Central (AE) / uaecentral
0.839 USD/h 4.54 USD/h
France South (FR) / francesouth
0.8708 USD/h 4.712 USD/h
Brazil South (BR) / brazilsouth
0.8715 USD/h 4.716 USD/h
Switzerland West (CH) / switzerlandwest
0.923 USD/h 4.9945 USD/h
Norway West (NO) / norwaywest
1.0513 USD/h 5.689 USD/h
Brazil Southeast (US) / brazilsoutheast
1.133 USD/h 6.131 USD/h

Processor

vCPUs
72
Hypervisor
Microsoft Hyper-V
CPU Allocation
Dedicated
CPU Cores
36
CPU Speed
2.6 GHz
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU Family
Xeon
CPU Model
8272CL
CPU L1D Cache
48 KiB
CPU L1D Cache Total
2 MiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
1 MiB
CPU L2 Cache
1 MiB
CPU L2 Cache Total
45 MiB
CPU L3 Cache
48 MiB
CPU L3 Cache Total
96 MiB
CPU Flags
fpu, vme, de, pse, tsc, msr, pae, mce, cx8, apic, sep, mtrr, pge, mca, cmov, pat, pse36, clflush, mmx, fxsr, sse, sse2, ss, ht, syscall, nx, pdpe1gb, rdtscp, lm, constant_tsc, rep_good, nopl, xtopology, cpuid, pni, pclmulqdq, vmx, ssse3, fma, cx16, pcid, sse4_1, sse4_2, movbe, popcnt, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, abm, 3dnowprefetch, pti, tpr_shadow, ept, vpid, fsgsbase, bmi1, hle, avx2, smep, bmi2, erms, invpcid, rtm, avx512f, avx512dq, rdseed, adx, smap, clflushopt, avx512cd, avx512bw, avx512vl, xsaveopt, xsavec, xsaves, vnmi, md_clear
Ecpus
46.1
Scalability
64.03

System Resources and Accelerators

MEMORY
Memory Amount
144 GiB
Memory Amount Actual
144 GiB
ACCELERATORS
Accelerator Count
0
Accelerator Memory Min
0 MiB
Accelerator Memory Total
0 MiB
Accelerators
    STORAGE
    Storage Size
    590 GB
    Storage Type
    ssd
    Storages
    • 590 GB ssd (temp disk)
    NETWORK
    Inbound Traffic
    0 GB/month
    Outbound Traffic
    0 GB/month
    IPv4
    0

    CPU and System Topology

    Server Description

    A high-density compute instance featuring seventy-two dedicated cores and strong multi-threaded performance for parallel processing workloads.

    Compute Optimized

    Microsoft Azure F72s_v2 is a compute-optimized virtual machine from the FSv2 family, powered by an Intel Xeon 8272CL processor on the x86_64 architecture. Operating under the Microsoft Hyper-V hypervisor, it delivers 72 dedicated vCPUs across 36 physical cores, 144.0 GB of RAM, and 590 GB of local storage. Benchmark results highlight strong multi-core CPU performance and high memory bandwidth, placing it in the top 25% tier for parallelized processing tasks. Conversely, single-core performance and LLM text generation speeds remain average to weak. This hardware profile provides high compute density, making it a cost-effective option for parallel workloads that do not demand high memory-to-CPU ratios. The F72s_v2 is well-suited for multi-threaded applications, software compilation, ray tracing, and high-throughput web serving.

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    2.5Score
    Precomputed compound score for Cache Intensive workloads. A weighted average (geometric mean) of benchmark scores compared to their medians: score = ∏ (x_i / m_i)^(w_i / Σw). The score of 1.0 represents a synthetic baseline server with the median performance of each component benchmark; 0.5 means roughly half the performance; and 2.0 means twice the performance of that reference profile. Component weights: 50% Redis RPS (pipeline=1, SET), 20% Redis RPS (pipeline=16, SET), 10% PassMark Memory Mark (composite), 10% Memory bandwidth (read, 16 MB ~ L3), 10% PassMark single-thread CPU. Rationale for component selection: In-memory key-value store workload, mixing direct Redis performance metrics with memory speed and latency benchmarks, and single-core CPU performance profiles.
    Component Score Weight Impact
    Raw Reference Normalized Target Actual
    Redis RPS (pipeline=1, SET) 6,468,370 ops/sec 1,919,289 ops/sec 3.4 50.00% 50.00% +84.00%
    Redis RPS (pipeline=16, SET) 36,945,788 ops/sec 11,728,058 ops/sec 3.1 20.00% 20.00% +25.00%
    PassMark Memory Mark (composite) 2,612 2,416 1.1 10.00% 10.00% +0.96%
    Memory bandwidth (read, 16 MB ~ L3) 220,404 MB/sec 108,466 MB/sec 2 10.00% 10.00% +7.20%
    PassMark single-thread CPU 2,114 Mops/s 2,365 Mops/s 0.88 10.00% 10.00% -1.30%

    Memory Bandwidth

    Compression

    OpenSSL

    Geekbench Single-Core

    Score: 1,252

    Geekbench Multi-Core

    Score: 12,846

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		 50,595
    Compression
    		 975,406
    Encryption
    		 30,480
    Extended Instructions
    		 62,396
    Floating Point Maths
    		 120,390
    Integer Maths
    		 224,679
    Physics
    		 4,409
    Prime Numbers
    		 260
    Single Threaded
    		 2,114
    String Sorting
    		 110,551

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		 2,612
    Database Operations
    		 21,218
    Memory Latency
    		 54
    Memory Read Cached
    		 24,631
    Memory Read Uncached
    		 10,153
    Memory Write
    		 9,032

    Stress-ng Stressors

    CPU integerSingleMultiScaling
    atomic
    		 192.88 220.93 1.15×
    bitops
    		 108,405 3,956,287 36.50×
    branch
    		 114,021 6,513,483 57.13×
    funccall
    		 526.55 19,457 36.95×
    funcret
    		 9,913 522,193 52.68×
    hash
    		 126,102 6,490,538 51.47×
    intmath
    		 50,567 2,228,077 44.06×
    nop
    		 155,596 4,943,132 31.77×
    prime
    		 330,856 12,900,410 38.99×
    rotate
    		 81,136 3,083,005 38.00×
    str
    		 749,319 28,168,390 37.59×
    wcs
    		 516,109 16,101,973 31.20×
    CryptographySingleMultiScaling
    crypt
    		 2,361 78,752 33.35×
    CompressionSingleMultiScaling
    jpeg
    		 593.83 26,003 43.79×
    zlib
    		 48.13 2,583 53.67×
    HypervisorSingleMultiScaling
    clock
    		 16,741 155,451 9.29×
    clone
    		 1,250 25,042 20.03×
    context
    		 1,745 80,236 45.98×
    cpu-sched
    		 80.67 4,571 56.67×
    eventfd
    		 35,401 1,576,138 44.52×
    fault
    		 11,304 148,198 13.11×
    fork
    		 4,317 47,630 11.03×
    getrandom
    		 39,873 1,655,352 41.52×
    hrtimers
    		 680,074 7,186,896 10.57×
    mprotect
    		 713,506 9,808,710 13.75×
    mtx
    		 13,415,847 1,383,464,913 103.12×
    munmap
    		 3,157 50,880 16.12×
    null
    		 187,371 8,163,646 43.57×
    pipe
    		 879,060 22,067,371 25.10×
    priv-instr
    		 397,976 10,738,303 26.98×
    resched
    		 6,852,485 4,239,259 0.62×
    switch
    		 69,898 12,232,793 175.01×
    timermix
    		 189,181 9,299,507 49.16×
    tlb-shootdown
    		 5,992 91,742 15.31×
    tsc
    		 744,333 36,734,406 49.35×
    usersyscall
    		 282,290 8,381,252 29.69×
    vdso
    		 75,763,645 3,153,208,922 41.62×
    vfork
    		 23,673 163,541 6.91×
    x86syscall
    		 2,082,634 91,323,492 43.85×
    yield
    		 34,309,531 38,409,343 1.12×
    zero
    		 1,568,171 66,644,786 42.50×
    CPU float/SIMDSingleMultiScaling
    eigen
    		 12,279 560,423 45.64×
    expmath
    		 2,400 124,380 51.83×
    fma
    		 564,201 35,094,333 62.20×
    fp
    		 165.6 6,122 36.97×
    fractal
    		 4.2 218.91 52.12×
    hyperbolic
    		 2,556 135,218 52.91×
    logmath
    		 4,017 209,257 52.10×
    matrix
    		 3,163 163,849 51.80×
    matrix-3d
    		 654.49 9,717 14.85×
    monte-carlo
    		 96.13 4,594 47.79×
    mpfr
    		 1,170 45,042 38.50×
    powmath
    		 1,401 74,996 53.53×
    trig
    		 1,641 74,465 45.39×
    veccmp
    		 17,262 585,650 33.93×
    vecfp
    		 1,269 56,026 44.15×
    vecmath
    		 2,724 138,585 50.87×
    vecshuf
    		 552.83 21,546 38.97×
    vecwide
    		 26,515 1,096,207 41.34×
    AlgorithmsSingleMultiScaling
    bitonicsort
    		 6.8 244.39 35.94×
    bsearch
    		 206.3 9,899 47.98×
    bubblesort
    		 0.8 14.4 18.00×
    fibsearch
    		 240.89 10,386 43.12×
    heapsort
    		 7.8 330.58 42.38×
    hsearch
    		 5,922 242,679 40.98×
    insertionsort
    		 8.6 187.2 21.77×
    malloc
    		 2,295,698 45,439,346 19.79×
    mcontend
    		 62.06 10,810 174.18×
    mergesort
    		 22.4 1,106 49.36×
    qsort
    		 17 643.38 37.85×
    skiplist
    		 4,511 237,186 52.58×
    tree
    		 1.4 43.19 30.85×
    MemorySingleMultiScaling
    cache
    		 319,537 1,633,871 5.11×
    cachehammer
    		 350,899 12,185,962 34.73×
    cacheline
    		 70.13 3,054 43.55×
    l1cache
    		 762.82 11,021 14.45×
    memcpy
    		 338.28 11,971 35.39×
    memrate
    		 1.76 26.16 14.86×
    memthrash
    		 144.96 1,057 7.29×
    ptr-chase
    		 41,844,680 737,173,658 17.62×
    spinmem
    		 318.58 1,593 5.00×
    stream
    		 19.72 23,622 1197.85×
    vm
    		 83,420 304,779 3.65×

    Stress-ng Raw Scores

    Stress-ng Relative Multicore Performance

    LLM Inference Speed for Prompt Processing

    LLM Inference Speed for Text Generation

    Static Web Server

    Redis

    Alternatives

    Servers of the Same Family

    INSTANCEvCPUsMEMORYAccelerators

    Similar Servers

    INSTANCEVENDORvCPUsMEMORYAccelerators

    F72s_v2 FAQs