D96a_v4 by Microsoft Azure

D96a_v4 is a DAv4 family [AMD processor] 96 vCPUs server offered by Microsoft Azure with 96 vCPUs, 384 GiB of memory and 2.58 TB of storage. The pricing starts at 0.5474 USD per hour.
96 vCPU
384 GiB Memory
2.58 TB Storage
Spare SCore
36,723
(All-cores)
782
(Single-core)

Specifications #

Server Metadata #

Vendor ID
azure
Server ID
Standard_D96a_v4
Name
D96a_v4
API Reference
Standard_D96a_v4
Description
DAv4 family [AMD processor] 96 vCPUs
Family
General Purpose
Series
Dav4
Hw Virt
Average Time To Start
33
Status
active
Observed At
2026-11-17T03:36:54.586457

Availability #

Region / IDSpotOndemand
Central India (IN) / centralindia
0.5474 USD/h
2.962 USD/h
Jio India Central (IN) / jioindiacentral
-
2.962 USD/h
Jio India West (IN) / jioindiawest
-
2.962 USD/h
India South Central (IN) / indiasouthcentral
0.8063 USD/h
4.363 USD/h
East US 2 (US) / eastus2
0.8516 USD/h
4.608 USD/h
North Central US (US) / northcentralus
0.8516 USD/h
4.608 USD/h
West US 2 (US) / westus2
0.8516 USD/h
4.608 USD/h
West US 3 (US) / westus3
0.8516 USD/h
4.608 USD/h
East US (US) / eastus
0.9718 USD/h
4.608 USD/h
Taiwan North (TW) / taiwannorth
0.8909 USD/h
4.821 USD/h
Spain Central (ES) / spaincentral
0.8959 USD/h
4.848 USD/h
Sweden Central (SE) / swedencentral
0.9048 USD/h
4.896 USD/h
Austria East (AT) / austriaeast
0.9225 USD/h
4.992 USD/h
Belgium Central (BE) / belgiumcentral
0.9225 USD/h
4.992 USD/h
Denmark East (DK) / denmarkeast
0.9225 USD/h
4.992 USD/h
Germany West Central (DE) / germanywestcentral
0.9225 USD/h
4.992 USD/h
Poland Central (PL) / polandcentral
0.9225 USD/h
4.992 USD/h
Mexico Central (ZA) / mexicocentral
0.9368 USD/h
5.069 USD/h
North Europe (IE) / northeurope
0.9491 USD/h
5.136 USD/h
Canada East (CA) / canadaeast
0.9522 USD/h
5.136 USD/h
Canada Central (CA) / canadacentral
0.9964 USD/h
5.136 USD/h
Indonesia Central (ID) / indonesiacentral
0.958 USD/h
5.184 USD/h
Malaysia West (MY) / malaysiawest
0.958 USD/h
5.184 USD/h
Central US (US) / centralus
0.9617 USD/h
5.207 USD/h
Korea South (KR) / koreasouth
0.9824 USD/h
5.316 USD/h
UK South (GB) / uksouth
0.9846 USD/h
5.328 USD/h
France Central (FR) / francecentral
0.9935 USD/h
5.376 USD/h
Israel Central (IL) / israelcentral
0.9935 USD/h
5.376 USD/h
Italy North (IT) / italynorth
0.9935 USD/h
5.376 USD/h
West US (US) / westus
0.9935 USD/h
5.376 USD/h
New Zealand North (NZ) / newzealandnorth
1.0059 USD/h
5.443 USD/h
West Europe (NL) / westeurope
1.1217 USD/h
5.52 USD/h
South Central US (US) / southcentralus
1.0219 USD/h
5.53 USD/h
West Central US (US) / westcentralus
1.0219 USD/h
5.53 USD/h
UK West (GB) / ukwest
1.0304 USD/h
5.576 USD/h
Qatar Central (QA) / qatarcentral
1.0441 USD/h
5.65 USD/h
UAE North (AE) / uaenorth
1.0441 USD/h
5.65 USD/h
Korea Central (KR) / koreacentral
1.0467 USD/h
5.664 USD/h
Australia Central 2 (AU) / australiacentral2
0.576 USD/h
5.76 USD/h
Australia Central (AU) / australiacentral
1.0644 USD/h
5.76 USD/h
Australia East (AU) / australiaeast
1.0644 USD/h
5.76 USD/h
Southeast Asia (SG) / southeastasia
1.0644 USD/h
5.76 USD/h
Chile Central (CL) / chilecentral
1.0729 USD/h
5.806 USD/h
West India (IN) / westindia
1.0899 USD/h
5.898 USD/h
Japan East (JP) / japaneast
1.0999 USD/h
5.952 USD/h
Japan West (JP) / japanwest
1.0999 USD/h
5.952 USD/h
Australia Southeast (AU) / australiasoutheast
1.107 USD/h
5.99 USD/h
South Africa North (ZA) / southafricanorth
1.1265 USD/h
6.096 USD/h
East Asia (HK) / eastasia
1.1709 USD/h
6.336 USD/h
Southeast Asia 3 (MY) / southeastasia3
1.1975 USD/h
6.48 USD/h
Germany North (DE) / germanynorth
1.1994 USD/h
6.49 USD/h
South India (IN) / southindia
1.2006 USD/h
6.497 USD/h
Norway East (NO) / norwayeast
1.2176 USD/h
6.589 USD/h
Switzerland North (CH) / switzerlandnorth
1.2176 USD/h
6.589 USD/h
France South (FR) / francesouth
1.2916 USD/h
6.989 USD/h
Brazil South (BR) / brazilsouth
1.3616 USD/h
7.344 USD/h
UAE Central (AE) / uaecentral
1.3607 USD/h
7.363 USD/h
South Africa West (ZA) / southafricawest
1.4645 USD/h
7.925 USD/h
Norway West (NO) / norwaywest
1.583 USD/h
8.566 USD/h
Switzerland West (CH) / switzerlandwest
1.583 USD/h
8.566 USD/h
Brazil Southeast (US) / brazilsoutheast
1.7643 USD/h
9.547 USD/h

Processor #

vCPUs
96
Hypervisor
Microsoft Hyper-V
CPU Allocation
Dedicated
CPU Cores
48
CPU Speed
3.35 GHz
CPU Architecture
x86_64
CPU Manufacturer
AMD
CPU Family
EPYC
CPU Model
7452
CPU L1D Cache
32 KiB
CPU L1D Cache Total
2 MiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
2 MiB
CPU L2 Cache
512 KiB
CPU L2 Cache Total
24 MiB
CPU L3 Cache
16 MiB
CPU L3 Cache Total
192 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, ht, syscall, nx, mmxext, fxsr_opt, pdpe1gb, rdtscp, lm, rep_good, nopl, cpuid, extd_apicid, pni, pclmulqdq, ssse3, fma, cx16, sse4_1, sse4_2, movbe, popcnt, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, cmp_legacy, cr8_legacy, abm, sse4a, misalignsse, 3dnowprefetch, osvw, topoext, ssbd, vmmcall, fsgsbase, bmi1, avx2, smep, bmi2, rdseed, adx, smap, clflushopt, clwb, sha_ni, xsaveopt, xsavec, xgetbv1, xsaves, xsaveerptr, arat, umip, rdpid
Ecpus
47
Scalability
48.96

System Resources and Accelerators #

Memory
Memory Amount
384 GiB
Memory Amount Actual
384 GiB
Accelerators
Accelerator Count
0
Accelerator Memory Min
0 MiB
Accelerator Memory Total
0 MiB
Accelerators
    Storage
    Storage Size
    2577 GB
    Storage Type
    ssd
    Storages
    • 2577 GB ssd (temp disk)
    Compatible Storage IDs
    StandardSSD_LRS, StandardSSD_ZRS, Standard_LRS
    Network
    Inbound Traffic
    0 GB/month
    Outbound Traffic
    0 GB/month
    IPv4
    0

    CPU and System Topology #

    Server Description #

    A high-core AMD EPYC virtual machine featuring 384 GB RAM and local SSD storage, optimized for parallelized and cryptographic workloads.

    Microsoft Azure D96a_v4 is a general-purpose virtual machine from the DAv4 family, powered by an AMD EPYC 7452 processor with 96 vCPUs and 48 physical cores. Operating at 3.35 GHz on the Microsoft Hyper-V hypervisor, it features 384.0 GB of RAM and 2458 GB of local SSD storage. While single-core benchmarks place it in the bottom 10%, its multi-threaded capabilities are robust, with software compilation, multi-threaded decompression, and PassMark CPU scores ranking in the top 25%. Cryptographic performance is exceptional, reaching the top 10% in encryption tests due to native AES and SHA_NI support. This instance is highly efficient for parallelized workloads, including multi-threaded software compilation, database operations, and medium-scale LLM inference, though it lacks GPU accelerators and complimentary public IPv4 addresses.

    Economics #

    Average Price per Region #

    Prices per Zone #

    Lowest Prices #

    Performance #

    Workload Profiles #

    2.6Score
    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) 7,079,093 ops/sec 1,933,252 ops/sec 3.7 50.00% 50.00% +92.00%
    Redis RPS (pipeline=16, SET) 43,844,427 ops/sec 11,814,424 ops/sec 3.7 20.00% 20.00% +30.00%
    PassMark Memory Mark (composite) 2,845 2,408 1.2 10.00% 10.00% +1.80%
    Memory bandwidth (read, 16 MB ~ L3) 161,978 MB/sec 108,674 MB/sec 1.5 10.00% 10.00% +4.10%
    PassMark single-thread CPU 2,179 Mops/s 2,365 Mops/s 0.92 10.00% 10.00% -0.83%

    Memory Bandwidth#

    Compression #

    OpenSSL #

    Geekbench Single-Core #

    Score: 1,399

    Geekbench Multi-Core #

    Score: 12,323

    Passmark CPU Scores #

    BENCHMARKSCORE
    Mark
    68,615
    Compression
    1,279,954
    Encryption
    96,990
    Extended Instructions
    71,637
    Floating Point Maths
    189,785
    Integer Maths
    319,278
    Physics
    7,478
    Prime Numbers
    462
    Single Threaded
    2,179
    String Sorting
    167,826

    Passmark Memory Scores #

    BENCHMARKSCORE
    Memory Mark
    2,845
    Database Operations
    21,125
    Memory Latency
    53
    Memory Read Cached
    24,897
    Memory Read Uncached
    12,598
    Memory Write
    10,801

    Stress-ng Stressors #

    CPU integerSingleMultiScaling
    atomic
    426.08 318.7 0.75×
    bitops
    122,053 6,661,513 54.58×
    branch
    152,732 12,111,918 79.30×
    funccall
    506.23 28,795 56.88×
    funcret
    14,951 939,608 62.85×
    hash
    132,588 9,239,681 69.69×
    intmath
    19,160 983,357 51.32×
    nop
    234,272 11,062,251 47.22×
    prime
    344,814 19,622,139 56.91×
    rotate
    59,881 2,915,040 48.68×
    str
    619,947 32,846,939 52.98×
    wcs
    456,772 22,572,098 49.42×
    CryptographySingleMultiScaling
    crypt
    2,818 131,146 46.53×
    CompressionSingleMultiScaling
    jpeg
    673.96 35,764 53.06×
    zlib
    46.47 3,970 85.43×
    HypervisorSingleMultiScaling
    clock
    3,095 32,148 10.39×
    clone
    700.53 21,870 31.22×
    context
    3,422 266,316 77.83×
    cpu-sched
    60.4 6,113 101.20×
    eventfd
    24,111 3,809,644 158.00×
    fault
    7,448 103,186 13.85×
    fork
    2,280 32,224 14.14×
    getrandom
    35,800 2,130,180 59.50×
    hrtimers
    66,142 1,490,603 22.54×
    mprotect
    284,551 12,486,130 43.88×
    mtx
    13,696,498 3,652,153,670 266.65×
    munmap
    1,614 32,458 20.11×
    null
    414,961 30,757,696 74.12×
    pipe
    1,686,061 29,591,688 17.55×
    priv-instr
    374,109 14,112,296 37.72×
    resched
    4,881,730 5,096,437 1.04×
    switch
    58,350 12,714,162 217.89×
    timermix
    41,723 2,645,929 63.42×
    tlb-shootdown
    8,053 86,829 10.78×
    tsc
    653,152 58,793,011 90.01×
    usersyscall
    293,060 12,102,211 41.30×
    vdso
    75,746,468 4,690,943,830 61.93×
    vfork
    9,345 80,004 8.56×
    x86syscall
    5,254,452 399,381,926 76.01×
    yield
    41,643,459 47,760,650 1.15×
    zero
    2,099,239 151,932,444 72.38×
    CPU float/SIMDSingleMultiScaling
    eigen
    12,424 658,143 52.97×
    expmath
    2,779 195,210 70.26×
    fma
    741,349 49,589,850 66.89×
    fp
    211.2 13,140 62.21×
    fractal
    6.4 415.69 64.95×
    hyperbolic
    3,676 243,360 66.20×
    logmath
    4,114 268,611 65.29×
    matrix
    2,535 208,852 82.38×
    matrix-3d
    66.15 4,584 69.30×
    monte-carlo
    182.88 13,167 72.00×
    mpfr
    1,279 70,310 54.97×
    powmath
    1,359 96,260 70.82×
    trig
    1,577 102,083 64.72×
    veccmp
    17,341 883,585 50.95×
    vecfp
    1,385 91,500 66.06×
    vecmath
    3,395 234,998 69.23×
    vecshuf
    489.26 26,708 54.59×
    vecwide
    26,810 1,284,076 47.89×
    AlgorithmsSingleMultiScaling
    bitonicsort
    7 367.98 52.57×
    bsearch
    227.74 11,172 49.06×
    bubblesort
    0.8 38.4 48.00×
    fibsearch
    263.08 15,218 57.85×
    heapsort
    8.6 506.17 58.86×
    hsearch
    5,393 278,226 51.59×
    insertionsort
    6.6 343.19 52.00×
    malloc
    2,418,308 66,973,761 27.69×
    mcontend
    41.48 8,533 205.72×
    mergesort
    25.8 1,627 63.05×
    qsort
    16.6 1,039 62.58×
    skiplist
    5,538 381,639 68.91×
    tree
    1.8 57.39 31.88×
    MemorySingleMultiScaling
    cache
    322,123 740,030 2.30×
    cachehammer
    242,432 7,748,110 31.96×
    cacheline
    34.16 2,920 85.49×
    l1cache
    196.84 8,652 43.95×
    memcpy
    357.87 15,393 43.01×
    memrate
    1.76 19.89 11.30×
    memthrash
    102.46 773.7 7.55×
    ptr-chase
    59,386,016 722,646,492 12.17×
    spinmem
    208.99 1,026 4.91×
    stream
    13.1 8,907 679.91×
    vm
    98,170 304,147 3.10×

    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
    D2a_v4 28 GiB0
    D4a_v4 416 GiB0
    D8a_v4 832 GiB0
    D16a_v4 1664 GiB0
    D32a_v4 32128 GiB0
    D48a_v4 48192 GiB0
    D64a_v4 64256 GiB0

    Similar Servers #

    INSTANCEVENDORvCPUsMEMORYAccelerators

    D96a_v4 FAQs #