B20ms by Microsoft Azure

B20ms is a BS family [Memory Intensive] [Premium Storage capable] [Intel processor] 20 vCPUs server offered by Microsoft Azure with 20 vCPUs, 80 GiB of memory and 164 GB of storage. The pricing starts at 0.832 USD per hour.
20 vCPU
80 GiB Memory
164 GB Storage
Spare SCore
24,697
(All-cores)
1,246
(Single-core)

Specifications

Server Metadata

Vendor ID
azure
Server ID
Standard_B20ms
Name
B20ms
API Reference
Standard_B20ms
Description
BS family [Memory Intensive] [Premium Storage capable] [Intel processor] 20 vCPUs
Family
BS
Hw Virt
Average Time To Start
36
Status
planned-for-retirement
Observed At
2026-11-01T13:17:13.688365

Availability

REGION / IDSPOTONDEMAND
East US (US) / eastus
- 0.832 USD/h
East US 2 (US) / eastus2
- 0.832 USD/h
North Central US (US) / northcentralus
- 0.832 USD/h
West US 2 (US) / westus2
- 0.832 USD/h
West US 3 (US) / westus3
- 0.832 USD/h
Sweden Central (SE) / swedencentral
- 0.864 USD/h
Central India (IN) / centralindia
- 0.896 USD/h
Jio India Central (IN) / jioindiacentral
- 0.896 USD/h
Jio India West (IN) / jioindiawest
- 0.896 USD/h
North Europe (IE) / northeurope
- 0.909 USD/h
Mexico Central (ZA) / mexicocentral
- 0.915 USD/h
Spain Central (ES) / spaincentral
- 0.915 USD/h
Canada Central (CA) / canadacentral
- 0.927 USD/h
Canada East (CA) / canadaeast
- 0.928 USD/h
UK West (GB) / ukwest
- 0.94 USD/h
France Central (FR) / francecentral
- 0.944 USD/h
UK South (GB) / uksouth
- 0.944 USD/h
Indonesia Central (ID) / indonesiacentral
- 0.95 USD/h
Malaysia West (MY) / malaysiawest
- 0.95 USD/h
Austria East (AT) / austriaeast
- 0.96 USD/h
Belgium Central (BE) / belgiumcentral
- 0.96 USD/h
Germany West Central (DE) / germanywestcentral
- 0.96 USD/h
Israel Central (IL) / israelcentral
- 0.96 USD/h
Italy North (IT) / italynorth
- 0.96 USD/h
Poland Central (PL) / polandcentral
- 0.96 USD/h
West Europe (NL) / westeurope
- 0.96 USD/h
West US (US) / westus
- 0.992 USD/h
Central US (US) / centralus
- 0.998 USD/h
South Central US (US) / southcentralus
- 0.998 USD/h
UAE North (AE) / uaenorth
- 0.998 USD/h
West Central US (US) / westcentralus
- 0.998 USD/h
Qatar Central (QA) / qatarcentral
- 1.003 USD/h
Korea South (KR) / koreasouth
- 1.037 USD/h
Korea Central (KR) / koreacentral
- 1.04 USD/h
Australia Central (AU) / australiacentral
- 1.056 USD/h
Australia Central 2 (AU) / australiacentral2
- 1.056 USD/h
Australia East (AU) / australiaeast
- 1.056 USD/h
Norway East (NO) / norwayeast
- 1.056 USD/h
Southeast Asia (SG) / southeastasia
- 1.056 USD/h
Switzerland North (CH) / switzerlandnorth
- 1.056 USD/h
Australia Southeast (AU) / australiasoutheast
- 1.057 USD/h
South Africa North (ZA) / southafricanorth
- 1.085 USD/h
Japan East (JP) / japaneast
- 1.088 USD/h
New Zealand North (NZ) / newzealandnorth
- 1.109 USD/h
Chile Central (CL) / chilecentral
- 1.165 USD/h
East Asia (HK) / eastasia
- 1.168 USD/h
South India (IN) / southindia
- 1.177 USD/h
West India (IN) / westindia
- 1.189 USD/h
Japan West (JP) / japanwest
- 1.198 USD/h
Germany North (DE) / germanynorth
- 1.248 USD/h
UAE Central (AE) / uaecentral
- 1.248 USD/h
Brazil South (BR) / brazilsouth
- 1.344 USD/h
France South (FR) / francesouth
- 1.352 USD/h
South Africa West (ZA) / southafricawest
- 1.3562 USD/h
Switzerland West (CH) / switzerlandwest
- 1.372 USD/h
Norway West (NO) / norwaywest
- 1.373 USD/h
Brazil Southeast (US) / brazilsoutheast
- 1.747 USD/h

Processor

vCPUs
20
Hypervisor
Microsoft Hyper-V
CPU Allocation
Burstable
CPU Cores
20
CPU Speed
2.6 GHz
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU Family
Xeon
CPU Model
8272CL
CPU L1D Cache
32 KiB
CPU L1D Cache Total
640 KiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
640 KiB
CPU L2 Cache
256 KiB
CPU L2 Cache Total
5 MiB
CPU L3 Cache
50 MiB
CPU L3 Cache Total
50 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, ssse3, fma, cx16, pcid, sse4_1, sse4_2, movbe, popcnt, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, abm, 3dnowprefetch, pti, fsgsbase, bmi1, hle, avx2, smep, bmi2, erms, invpcid, rtm, avx512f, avx512dq, rdseed, adx, smap, clflushopt, avx512cd, avx512bw, avx512vl, xsaveopt, xsavec, xsaves, md_clear
Ecpus
19.8
Scalability
99

System Resources and Accelerators

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

    CPU and System Topology

    Server Description

    A burstable, memory-dense virtual machine featuring twenty cores and eighty gigabytes of RAM for variable-demand database and web serving workloads.

    Burstable & BudgetGeneral Purpose

    Microsoft Azure B20ms is a burstable virtual machine from the BS family, configured with 20 vCPUs and 80.0 GB of RAM. Powered by an Intel Xeon 8272CL x86_64 processor running at 2.6 GHz, this instance features 20 physical cores, a 50 MB L3 cache, and 164 GB of local storage. Operating under the Microsoft Hyper-V hypervisor, it lacks dedicated GPUs. Benchmark data reveals a clear tradeoff: single-core performance, single-threaded compression, and text processing fall into the bottom 25%, while multi-core execution, memory bandwidth, and database operations consistently achieve average tiering. This resource profile provides qualitative cost efficiency for memory-intensive, variable-demand workloads. It is suited for database operations, software compilation, and static web serving, but is less effective for sustained compute-heavy tasks or real-time LLM text generation.

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    1.1Score
    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) 2,549,248 ops/sec 1,918,343 ops/sec 1.3 50.00% 50.00% +14.00%
    Redis RPS (pipeline=16, SET) 16,071,944 ops/sec 11,704,734 ops/sec 1.3 20.00% 20.00% +5.40%
    PassMark Memory Mark (composite) 2,339 2,416 0.96 10.00% 10.00% -0.41%
    Memory bandwidth (read, 16 MB ~ L3) 112,339 MB/sec 108,466 MB/sec 1 10.00% 10.00% 0.00%
    PassMark single-thread CPU 1,571 Mops/s 2,365 Mops/s 0.67 10.00% 10.00% -3.90%

    Memory Bandwidth

    Compression

    OpenSSL

    Geekbench Single-Core

    Score: 1,093

    Geekbench Multi-Core

    Score: 9,962

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		 19,838
    Compression
    		 291,983
    Encryption
    		 7,418
    Extended Instructions
    		 24,848
    Floating Point Maths
    		 48,349
    Integer Maths
    		 58,022
    Physics
    		 2,270
    Prime Numbers
    		 139
    Single Threaded
    		 1,571
    String Sorting
    		 34,762

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		 2,339
    Database Operations
    		 10,959
    Memory Latency
    		 56
    Memory Read Cached
    		 18,576
    Memory Read Uncached
    		 9,769
    Memory Write
    		 8,850

    Stress-ng Stressors

    CPU integerSingleMultiScaling
    atomic
    		 185.97 210.72 1.13×
    bitops
    		 85,028 1,670,022 19.64×
    branch
    		 109,051 2,163,027 19.83×
    funccall
    		 373.39 7,441 19.93×
    funcret
    		 7,244 145,111 20.03×
    hash
    		 89,658 1,742,192 19.43×
    intmath
    		 14,721 294,476 20.00×
    nop
    		 124,246 2,524,624 20.32×
    prime
    		 219,838 4,322,451 19.66×
    rotate
    		 68,667 1,372,927 19.99×
    str
    		 464,002 9,574,996 20.64×
    wcs
    		 310,677 5,875,856 18.91×
    CryptographySingleMultiScaling
    crypt
    		 1,660 31,910 19.22×
    CompressionSingleMultiScaling
    jpeg
    		 367.47 7,382 20.09×
    zlib
    		 48.27 651.46 13.50×
    HypervisorSingleMultiScaling
    clock
    		 12,431 109,988 8.85×
    clone
    		 979.87 10,046 10.25×
    context
    		 1,396 20,586 14.75×
    cpu-sched
    		 49.02 1,269 25.88×
    eventfd
    		 20,113 340,356 16.92×
    fault
    		 4,612 68,641 14.88×
    fork
    		 2,162 21,129 9.77×
    getrandom
    		 30,860 596,947 19.34×
    hrtimers
    		 268,671 2,582,888 9.61×
    mprotect
    		 800,363 3,215,510 4.02×
    mtx
    		 11,563,849 714,632,525 61.80×
    munmap
    		 1,498 19,638 13.11×
    null
    		 142,942 2,121,835 14.84×
    pipe
    		 659,551 5,612,808 8.51×
    priv-instr
    		 316,422 5,900,889 18.65×
    resched
    		 3,906,783 1,743,786 0.45×
    switch
    		 67,732 3,375,403 49.83×
    timermix
    		 130,506 2,499,536 19.15×
    tlb-shootdown
    		 7,446 41,911 5.63×
    tsc
    		 702,175 14,258,112 20.31×
    usersyscall
    		 222,425 3,945,952 17.74×
    vdso
    		 54,472,606 1,074,416,891 19.72×
    vfork
    		 8,524 98,129 11.51×
    x86syscall
    		 1,669,898 22,285,508 13.35×
    yield
    		 9,741,780 10,212,729 1.05×
    zero
    		 1,093,204 18,197,655 16.65×
    CPU float/SIMDSingleMultiScaling
    eigen
    		 10,380 206,121 19.86×
    expmath
    		 1,861 35,999 19.34×
    fma
    		 420,840 8,220,761 19.53×
    fp
    		 88.8 1,747 19.67×
    fractal
    		 3.6 70.21 19.50×
    hyperbolic
    		 2,043 39,827 19.50×
    logmath
    		 3,155 63,821 20.23×
    matrix
    		 2,652 48,770 18.39×
    matrix-3d
    		 567.36 1,954 3.44×
    monte-carlo
    		 75.23 1,267 16.85×
    mpfr
    		 814.32 15,899 19.52×
    powmath
    		 1,095 22,189 20.26×
    trig
    		 1,280 25,724 20.10×
    veccmp
    		 9,408 187,518 19.93×
    vecfp
    		 606.46 11,955 19.71×
    vecmath
    		 1,568 30,364 19.37×
    vecshuf
    		 387.57 7,732 19.95×
    vecwide
    		 17,311 335,684 19.39×
    AlgorithmsSingleMultiScaling
    bitonicsort
    		 5 97.6 19.52×
    bsearch
    		 159.53 3,162 19.82×
    bubblesort
    		 0.4 8 20.00×
    fibsearch
    		 186.44 3,667 19.67×
    heapsort
    		 5.6 110.8 19.79×
    hsearch
    		 3,034 58,538 19.29×
    insertionsort
    		 5.6 108 19.29×
    malloc
    		 1,338,331 26,405,763 19.73×
    mcontend
    		 113.89 4,228 37.12×
    mergesort
    		 17.4 349.19 20.07×
    qsort
    		 12.4 251.79 20.31×
    skiplist
    		 4,053 81,048 20.00×
    tree
    		 1 12 12.00×
    MemorySingleMultiScaling
    cache
    		 158,204 424,134 2.68×
    cachehammer
    		 413,986 7,296,972 17.63×
    cacheline
    		 22.96 475.05 20.69×
    l1cache
    		 134.17 2,622 19.54×
    memcpy
    		 301.43 5,816 19.29×
    memrate
    		 0.97 7.58 7.81×
    memthrash
    		 74.71 426.29 5.71×
    ptr-chase
    		 18,628,364 178,118,761 9.56×
    spinmem
    		 138.19 - -
    stream
    		 23.6 2,072 87.81×
    vm
    		 33,185 293,917 8.86×

    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

    B20ms FAQs