M96-48bds_2_v3 by Microsoft Azure

M96-48bds_2_v3 is a MBDSMediumMemoryv3 family [Block Storage performance] [Local Disk] [Premium Storage capable] [Intel processor] 96 vCPUs server offered by Microsoft Azure with 48 vCPUs, 1946 GiB of memory and 4.61 TB of storage. The pricing starts at 6.894 USD per hour.
48 vCPU
1946 GiB Memory
4.61 TB Storage
Spare SCore
50,642
(All-cores)
2,058
(Single-core)

Specifications

Server Metadata

Vendor ID
azure
Server ID
Standard_M96-48bds_2_v3
Name
M96-48bds_2_v3
API Reference
Standard_M96-48bds_2_v3
Description
MBDSMediumMemoryv3 family [Block Storage performance] [Local Disk] [Premium Storage capable] [Intel processor] 96 vCPUs
Family
MBDSMediumMemoryv3
Hw Virt
Average Time To Start
74
Status
active
Observed At
2026-10-24T11:47:00.105710

Availability

REGION / IDSPOTONDEMAND
North Central US (US) / northcentralus
7.7346 USD/h 34.4678 USD/h
West US 3 (US) / westus3
7.7346 USD/h 34.4678 USD/h
India South Central (IN) / indiasouthcentral
6.894 USD/h 34.47 USD/h
Central India (IN) / centralindia
7.7351 USD/h 34.47 USD/h
East US (US) / eastus
7.7351 USD/h 34.47 USD/h
East US 2 (US) / eastus2
7.7351 USD/h 34.47 USD/h
South India (IN) / southindia
7.7351 USD/h 34.47 USD/h
West US 2 (US) / westus2
7.7351 USD/h 34.47 USD/h
Jio India Central (IN) / jioindiacentral
- 34.47 USD/h
Jio India West (IN) / jioindiawest
- 34.47 USD/h
Central US (US) / centralus
7.1979 USD/h 38.9499 USD/h
Malaysia West (MY) / malaysiawest
9.05 USD/h 40.3299 USD/h
Indonesia Central (ID) / indonesiacentral
9.0501 USD/h 40.33 USD/h
Australia East (AU) / australiaeast
9.2698 USD/h 41.3093 USD/h
North Europe (IE) / northeurope
9.2812 USD/h 41.36 USD/h
South Central US (US) / southcentralus
9.2812 USD/h 41.36 USD/h
Canada Central (CA) / canadacentral
10.0473 USD/h 41.364 USD/h
Austria East (AT) / austriaeast
9.3306 USD/h 41.58 USD/h
Germany West Central (DE) / germanywestcentral
9.3312 USD/h 41.5829 USD/h
UK South (GB) / uksouth
9.6688 USD/h 43.0875 USD/h
New Zealand North (NZ) / newzealandnorth
9.7333 USD/h 43.3747 USD/h
Southeast Asia (SG) / southeastasia
10.055 USD/h 44.8082 USD/h
West Europe (NL) / westeurope
10.0554 USD/h 44.81 USD/h
Korea Central (KR) / koreacentral
10.4423 USD/h 46.5345 USD/h
Sweden Central (SE) / swedencentral
10.8287 USD/h 48.2563 USD/h
Switzerland North (CH) / switzerlandnorth
11.0611 USD/h 49.2921 USD/h
UAE North (AE) / uaenorth
11.0611 USD/h 49.2921 USD/h
Japan East (JP) / japaneast
11.2158 USD/h 49.9815 USD/h
East Asia (HK) / eastasia
11.9894 USD/h 53.4285 USD/h
Japan West (JP) / japanwest
11.9894 USD/h 53.4285 USD/h
Brazil South (BR) / brazilsouth
15.4701 USD/h 68.94 USD/h

Processor

vCPUs
48
Hypervisor
Microsoft Hyper-V
CPU Allocation
Dedicated
CPU Cores
24
CPU Speed
4 GHz
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU Family
Xeon
CPU Model
8473C
CPU L1D Cache
48 KiB
CPU L1D Cache Total
1 MiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
768 KiB
CPU L2 Cache
2 MiB
CPU L2 Cache Total
48 MiB
CPU L3 Cache
105 MiB
CPU L3 Cache Total
210 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, tsc_reliable, nonstop_tsc, cpuid, aperfmperf, pni, pclmulqdq, ssse3, fma, cx16, pcid, sse4_1, sse4_2, x2apic, movbe, popcnt, tsc_deadline_timer, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, abm, 3dnowprefetch, fsgsbase, tsc_adjust, bmi1, hle, avx2, smep, bmi2, erms, invpcid, rtm, avx512f, avx512dq, rdseed, adx, smap, avx512ifma, clflushopt, clwb, avx512cd, sha_ni, avx512bw, avx512vl, xsaveopt, xsavec, xgetbv1, xsaves, user_shstk, avx_vnni, avx512_bf16, avx512vbmi, umip, waitpkg, avx512_vbmi2, gfni, vaes, vpclmulqdq, avx512_vnni, avx512_bitalg, avx512_vpopcntdq, la57, rdpid, cldemote, movdiri, movdir64b, fsrm, serialize, tsxldtrk, ibt, amx_bf16, avx512_fp16, amx_tile, amx_int8, arch_capabilities
Ecpus
24.6
Scalability
51.25

System Resources and Accelerators

MEMORY
Memory Amount
2 TB
Memory Amount Actual
2 TB
ACCELERATORS
Accelerator Count
0
Accelerator Memory Min
0 MiB
Accelerator Memory Total
0 MiB
Accelerators
    STORAGE
    Storage Size
    4608 GB
    Storage Type
    ssd
    Storages
    • 4608 GB ssd (temp disk)
    NETWORK
    Inbound Traffic
    0 GB/month
    Outbound Traffic
    0 GB/month
    IPv4
    0

    CPU and System Topology

    Server Description

    An x86_64 server combining 1.9 TB of system memory, a 4.6 TB local SSD, and dedicated Intel Xeon processing.

    Memory OptimizedStorage & DatabaseCompute Optimized

    Microsoft Azure M96-48bds_2_v3 is an x86_64 memory-optimized server in the MBDSMediumMemoryv3 family. Powered by an Intel Xeon 8473C processor running at 4.0 GHz, it features 24 physical cores, 48 dedicated vCPUs, and 1946.0 GB of RAM, providing 40.54 GB of memory per core. The instance includes a 4608 GB local SSD and lacks GPU acceleration. Benchmark data shows strong memory bandwidth, with small block reads at 4147.11 GB/sec, and high database performance, including 66.02 million Redis SET operations per second. The processor supports advanced instructions like AVX-512 and AMX, aiding in tasks like software compilation and LLM prompt processing. This server is designed for memory-intensive workloads, large databases, and in-memory caching.

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    3.3Score
    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) 9,196,543 ops/sec 1,919,289 ops/sec 4.8 50.00% 50.00% +120.00%
    Redis RPS (pipeline=16, SET) 45,653,803 ops/sec 11,728,058 ops/sec 3.8 20.00% 20.00% +31.00%
    PassMark Memory Mark (composite) 2,426 2,416 1 10.00% 10.00% 0.00%
    Memory bandwidth (read, 16 MB ~ L3) 326,396 MB/sec 108,466 MB/sec 3 10.00% 10.00% +12.00%
    PassMark single-thread CPU 3,166 Mops/s 2,365 Mops/s 1.3 10.00% 10.00% +2.70%

    Memory Bandwidth

    Compression

    OpenSSL

    Geekbench Single-Core

    Score: 2,205

    Geekbench Multi-Core

    Score: 19,086

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		 67,646
    Compression
    		 889,619
    Encryption
    		 53,326
    Extended Instructions
    		 69,362
    Floating Point Maths
    		 172,726
    Integer Maths
    		 242,567
    Physics
    		 9,673
    Prime Numbers
    		 488
    Single Threaded
    		 3,166
    String Sorting
    		 121,843

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		 2,426
    Database Operations
    		 26,777
    Memory Latency
    		 69
    Memory Read Cached
    		 28,400
    Memory Read Uncached
    		 10,309
    Memory Write
    		 9,026

    Stress-ng Stressors

    CPU integerSingleMultiScaling
    atomic
    		 209.14 126.89 0.61×
    bitops
    		 156,649 4,027,480 25.71×
    branch
    		 142,974 4,906,098 34.31×
    funccall
    		 762.35 23,290 30.55×
    funcret
    		 13,152 471,639 35.86×
    hash
    		 184,601 6,116,575 33.13×
    intmath
    		 68,318 1,777,630 26.02×
    nop
    		 276,682 4,912,597 17.76×
    prime
    		 495,214 12,855,429 25.96×
    rotate
    		 104,792 2,472,639 23.60×
    str
    		 997,343 30,496,324 30.58×
    wcs
    		 729,401 19,377,234 26.57×
    CryptographySingleMultiScaling
    crypt
    		 3,534 74,860 21.18×
    CompressionSingleMultiScaling
    jpeg
    		 886.92 25,276 28.50×
    zlib
    		 59.58 2,235 37.51×
    HypervisorSingleMultiScaling
    clock
    		 33,106 173,373 5.24×
    clone
    		 1,291 22,056 17.09×
    context
    		 5,580 203,388 36.45×
    cpu-sched
    		 61 3,026 49.60×
    eventfd
    		 36,018 1,512,769 42.00×
    fault
    		 24,647 178,102 7.23×
    fork
    		 5,551 50,543 9.11×
    getrandom
    		 62,206 2,078,082 33.41×
    hrtimers
    		 174,241 4,548,401 26.10×
    mprotect
    		 554,371 13,236,322 23.88×
    mtx
    		 17,619,487 1,226,106,975 69.59×
    munmap
    		 4,219 51,884 12.30×
    null
    		 900,439 31,455,368 34.93×
    pipe
    		 652,300 24,715,396 37.89×
    priv-instr
    		 899,540 12,494,216 13.89×
    resched
    		 10,373,212 3,789,123 0.37×
    switch
    		 52,980 13,988,154 264.03×
    timermix
    		 247,164 8,707,259 35.23×
    tlb-shootdown
    		 4,621 69,303 15.00×
    tsc
    		 953,892 30,742,591 32.23×
    usersyscall
    		 729,127 12,087,095 16.58×
    vdso
    		 109,505,484 3,582,782,357 32.72×
    vfork
    		 25,440 157,456 6.19×
    x86syscall
    		 10,774,486 399,655,098 37.09×
    yield
    		 58,437,197 51,949,413 0.89×
    zero
    		 6,032,423 215,911,346 35.79×
    CPU float/SIMDSingleMultiScaling
    eigen
    		 16,951 475,020 28.02×
    expmath
    		 3,691 117,019 31.70×
    fma
    		 1,419,494 44,914,748 31.64×
    fp
    		 323.56 11,187 34.57×
    fractal
    		 7.2 234.3 32.54×
    hyperbolic
    		 4,210 139,572 33.16×
    logmath
    		 5,524 196,385 35.55×
    matrix
    		 6,920 163,749 23.66×
    matrix-3d
    		 665.74 9,926 14.91×
    monte-carlo
    		 312.67 11,286 36.10×
    mpfr
    		 1,830 45,269 24.73×
    powmath
    		 2,060 68,512 33.25×
    trig
    		 2,785 79,986 28.72×
    veccmp
    		 26,984 649,551 24.07×
    vecfp
    		 2,416 60,342 24.98×
    vecmath
    		 4,038 128,883 31.92×
    vecshuf
    		 3,964 166,692 42.05×
    vecwide
    		 83,526 2,346,281 28.09×
    AlgorithmsSingleMultiScaling
    bitonicsort
    		 9.2 240.99 26.19×
    bsearch
    		 303.84 8,587 28.26×
    bubblesort
    		 1 28.8 28.80×
    fibsearch
    		 323.78 8,754 27.04×
    heapsort
    		 10.4 323.99 31.15×
    hsearch
    		 8,942 237,863 26.60×
    insertionsort
    		 13.4 268 20.00×
    malloc
    		 2,852,130 48,759,076 17.10×
    mcontend
    		 56.3 7,635 135.62×
    mergesort
    		 32.6 1,027 31.51×
    qsort
    		 23 705.78 30.69×
    skiplist
    		 5,723 194,757 34.03×
    tree
    		 1.6 47 29.38×
    MemorySingleMultiScaling
    cache
    		 216,964 2,220,025 10.23×
    cachehammer
    		 213,637 5,808,448 27.19×
    cacheline
    		 20.58 2,352 114.29×
    l1cache
    		 1,162 10,410 8.96×
    memcpy
    		 467.04 10,291 22.03×
    memrate
    		 0.17 7.4 43.53×
    memthrash
    		 111.96 1,130 10.09×
    ptr-chase
    		 32,106,928 490,982,239 15.29×
    spinmem
    		 281.18 1,366 4.86×
    stream
    		 7.12 6,827 958.81×
    vm
    		 166,509 299,569 1.80×

    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
    M16bds_v3 16128 GiB0
    M32bds_v3 32256 GiB0
    M48bds_v3 48384 GiB0
    M64bds_v3 64512 GiB0
    M128-64bds_v3 641024 GiB0
    M64bds_1_v3 641397 GiB0
    M128-64bds_3_v3 642794 GiB0

    Similar Servers

    INSTANCEVENDORvCPUsMEMORYAccelerators

    M96-48bds_2_v3 FAQs