D48pds_v5 by Microsoft Azure

D48pds_v5 is a DPDSv5 family [ARM processor] [Local Disk] [Premium Storage capable] 48 vCPUs server offered by Microsoft Azure with 48 vCPUs, 192 GiB of memory and 1.84 TB of storage. The pricing starts at 0.268 USD per hour.
48 vCPU
192 GiB Memory
1.84 TB Storage
Spare SCore
85,111
(All-cores)
1,822
(Single-core)

Specifications

Server Metadata

Vendor ID
azure
Server ID
Standard_D48pds_v5
Name
D48pds_v5
API Reference
Standard_D48pds_v5
Description
DPDSv5 family [ARM processor] [Local Disk] [Premium Storage capable] 48 vCPUs
Family
DPDSv5
Hw Virt
Average Time To Start
29
Status
active
Observed At
2026-10-24T15:35:47.612341

Availability

REGION / IDSPOTONDEMAND
Central India (IN) / centralindia
0.268 USD/h 1.45 USD/h
Jio India Central (IN) / jioindiacentral
- 1.45 USD/h
Jio India West (IN) / jioindiawest
- 1.45 USD/h
East US 2 (US) / eastus2
0.401 USD/h 2.17 USD/h
North Central US (US) / northcentralus
0.401 USD/h 2.17 USD/h
West US 2 (US) / westus2
0.401 USD/h 2.17 USD/h
West US 3 (US) / westus3
0.401 USD/h 2.17 USD/h
East US (US) / eastus
0.4587 USD/h 2.17 USD/h
Sweden Central (SE) / swedencentral
0.4258 USD/h 2.304 USD/h
North Europe (IE) / northeurope
0.447 USD/h 2.419 USD/h
Canada Central (CA) / canadacentral
0.448 USD/h 2.424 USD/h
Central US (US) / centralus
0.4531 USD/h 2.453 USD/h
UK South (GB) / uksouth
0.3994 USD/h 2.501 USD/h
France Central (FR) / francecentral
0.4701 USD/h 2.544 USD/h
West US (US) / westus
0.4701 USD/h 2.544 USD/h
South Central US (US) / southcentralus
0.4808 USD/h 2.602 USD/h
West Central US (US) / westcentralus
0.4812 USD/h 2.604 USD/h
Germany West Central (DE) / germanywestcentral
0.4834 USD/h 2.616 USD/h
West Europe (NL) / westeurope
0.4834 USD/h 2.616 USD/h
Southeast Asia (SG) / southeastasia
0.5012 USD/h 2.712 USD/h
Australia East (AU) / australiaeast
0.5056 USD/h 2.736 USD/h
Japan East (JP) / japaneast
0.5189 USD/h 2.808 USD/h
Switzerland North (CH) / switzerlandnorth
0.5322 USD/h 2.88 USD/h
East Asia (HK) / eastasia
0.5522 USD/h 2.988 USD/h

Processor

vCPUs
48
Hypervisor
Microsoft Hyper-V
CPU Allocation
Dedicated
CPU Cores
48
CPU Speed
3 GHz
CPU Architecture
arm64
CPU Manufacturer
Ampere
CPU Family
ARMv8
CPU Model
Neoverse-N1
CPU L1D Cache
64 KiB
CPU L1D Cache Total
3 MiB
CPU L1I Cache
64 KiB
CPU L1I Cache Total
3 MiB
CPU L2 Cache
1 MiB
CPU L2 Cache Total
48 MiB
CPU L3 Cache
32 MiB
CPU L3 Cache Total
32 MiB
CPU Flags
fp, asimd, evtstrm, aes, pmull, sha1, sha2, crc32, atomics, fphp, asimdhp, cpuid, asimdrdm, lrcpc, dcpop, asimddp
Ecpus
46.7
Scalability
97.29

System Resources and Accelerators

MEMORY
Memory Amount
192 GiB
Memory Amount Actual
192 GiB
ACCELERATORS
Accelerator Count
0
Accelerator Memory Min
0 MiB
Accelerator Memory Total
0 MiB
Accelerators
    STORAGE
    Storage Size
    1843 GB
    Storage Type
    ssd
    Storages
    • 1843 GB ssd (temp disk)
    NETWORK
    Inbound Traffic
    0 GB/month
    Outbound Traffic
    0 GB/month
    IPv4
    0

    CPU and System Topology

    Server Description

    An ARM64 instance with 48 dedicated cores, 192 GB RAM, and local SSD storage optimized for multi-threaded workloads.

    General PurposeStorage & Database

    The Microsoft Azure D48pds_v5 is a general-purpose instance driven by a 48-core Ampere Altra Neoverse-N1 ARM64 processor operating at 3.0 GHz. It offers 192 GB of system memory along with a bundled 1,843 GB local SSD, establishing a balanced resource allocation profile. Benchmarks demonstrate strong multi-threaded performance, notably in multi-core stress testing, Clang software compilation, and Redis SET operations exceeding 60 million ops/sec. However, single-core speeds and bzip3 decompression metrics remain lower relative to x86 alternatives, highlighting a tradeoff between single-thread latency and scalable multi-core density. Without hardware hyperthreading, each core delivers predictable execution suitable for parallelized microservices, build farms, caching layers, and database workloads utilizing high-speed local storage.

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    2.2Score
    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) 6,210,735 ops/sec 1,919,289 ops/sec 3.3 50.00% 50.00% +82.00%
    Redis RPS (pipeline=16, SET) 35,542,012 ops/sec 11,728,058 ops/sec 3 20.00% 20.00% +25.00%
    PassMark Memory Mark (composite) 1,985 2,416 0.83 10.00% 10.00% -1.80%
    Memory bandwidth (read, 16 MB ~ L3) 187,335 MB/sec 108,466 MB/sec 1.7 10.00% 10.00% +5.40%
    PassMark single-thread CPU 1,333 Mops/s 2,365 Mops/s 0.54 10.00% 10.00% -6.00%

    Memory Bandwidth

    Compression

    OpenSSL

    Geekbench Single-Core

    Score: 1,121

    Geekbench Multi-Core

    Score: 14,198

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		 27,156
    Compression
    		 266,568
    Encryption
    		 7,672
    Extended Instructions
    		 34,969
    Floating Point Maths
    		 168,308
    Integer Maths
    		 221,115
    Physics
    		 3,556
    Prime Numbers
    		 213
    Single Threaded
    		 1,333
    String Sorting
    		 154,041

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		 1,985
    Database Operations
    		 21,206
    Memory Latency
    		 70
    Memory Read Cached
    		 11,381
    Memory Read Uncached
    		 8,988
    Memory Write
    		 15,062

    Stress-ng Stressors

    CPU integerSingleMultiScaling
    atomic
    		 97.71 25.18 0.26×
    bitops
    		 108,442 5,200,553 47.96×
    branch
    		 220,872 10,592,344 47.96×
    funccall
    		 1,337 64,130 47.96×
    funcret
    		 20,938 997,137 47.62×
    hash
    		 128,224 6,136,318 47.86×
    intmath
    		 30,955 1,506,463 48.67×
    nop
    		 173,517 8,315,357 47.92×
    prime
    		 336,813 16,151,252 47.95×
    rotate
    		 96,526 4,627,201 47.94×
    str
    		 514,705 24,668,587 47.93×
    wcs
    		 119,277 5,713,193 47.90×
    CryptographySingleMultiScaling
    crypt
    		 2,893 139,458 48.20×
    CompressionSingleMultiScaling
    jpeg
    		 521.64 24,351 46.68×
    zlib
    		 36.39 1,718 47.22×
    HypervisorSingleMultiScaling
    clock
    		 22,325 237,023 10.62×
    clone
    		 209.1 6,808 32.56×
    context
    		 2,617 124,894 47.73×
    cpu-sched
    		 60.53 3,111 51.39×
    eventfd
    		 51,854 1,272,577 24.54×
    fault
    		 13,426 315,926 23.53×
    fork
    		 3,618 59,345 16.40×
    getrandom
    		 49,125 2,354,769 47.93×
    hrtimers
    		 129,376 5,715,896 44.18×
    mprotect
    		 866,366 14,995,923 17.31×
    mtx
    		 8,171,960 1,601,998,534 196.04×
    munmap
    		 2,580 68,015 26.36×
    null
    		 463,015 22,202,637 47.95×
    pipe
    		 627,177 20,265,706 32.31×
    priv-instr
    		 731,453 12,715,520 17.38×
    resched
    		 5,675,546 7,787,874 1.37×
    switch
    		 78,045 14,757,956 189.09×
    timermix
    		 404,813 19,322,082 47.73×
    tlb-shootdown
    		 4,916 5,273 1.07×
    vdso
    		 46,818,091 2,244,035,414 47.93×
    vfork
    		 23,103 245,660 10.63×
    yield
    		 41,486,456 50,177,152 1.21×
    zero
    		 2,078,435 99,441,577 47.84×
    CPU float/SIMDSingleMultiScaling
    eigen
    		 1,336 64,072 47.95×
    expmath
    		 240.08 11,509 47.94×
    fma
    		 640,751 30,732,039 47.96×
    fp
    		 103.2 4,965 48.11×
    fractal
    		 6.6 308.91 46.80×
    hyperbolic
    		 244.28 11,712 47.95×
    logmath
    		 293.59 14,079 47.96×
    matrix
    		 5,349 256,454 47.94×
    matrix-3d
    		 512.45 5,869 11.45×
    monte-carlo
    		 193.06 9,264 47.99×
    mpfr
    		 641.87 30,797 47.98×
    powmath
    		 176.52 8,464 47.95×
    trig
    		 329.61 15,802 47.94×
    veccmp
    		 17,165 822,329 47.91×
    vecfp
    		 745.46 35,728 47.93×
    vecmath
    		 3,090 148,150 47.94×
    vecshuf
    		 673.4 32,216 47.84×
    vecwide
    		 18,421 882,775 47.92×
    AlgorithmsSingleMultiScaling
    bitonicsort
    		 5.4 259.18 48.00×
    bsearch
    		 258.34 12,387 47.95×
    bubblesort
    		 0.8 37.8 47.25×
    fibsearch
    		 273.29 13,120 48.01×
    heapsort
    		 8.2 394.98 48.17×
    hsearch
    		 4,928 235,720 47.84×
    insertionsort
    		 4.6 220.4 47.91×
    malloc
    		 2,787,114 71,882,250 25.79×
    mcontend
    		 196.14 100,163 510.67×
    mergesort
    		 28 1,290 46.07×
    qsort
    		 19.4 935.74 48.23×
    skiplist
    		 5,357 256,525 47.89×
    tree
    		 1.2 28.74 23.95×
    MemorySingleMultiScaling
    cache
    		 363,606 643,382 1.77×
    cachehammer
    		 144,835 430,592 2.97×
    cacheline
    		 21.78 2,785 127.85×
    l1cache
    		 182.16 8,601 47.21×
    memcpy
    		 443.74 21,240 47.87×
    memrate
    		 3.13 29.13 9.31×
    memthrash
    		 196.15 344.15 1.75×
    ptr-chase
    		 13,277,317 410,022,541 30.88×
    spinmem
    		 899.97 1,831 2.04×
    stream
    		 69.98 53,179 759.91×
    vm
    		 60,043 2,435,290 40.56×

    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
    D2pds_v5 28 GiB0
    D4pds_v5 416 GiB0
    D8pds_v5 832 GiB0
    D16pds_v5 1664 GiB0
    D32pds_v5 32128 GiB0
    D64pds_v5 64208 GiB0

    Similar Servers

    INSTANCEVENDORvCPUsMEMORYAccelerators

    D48pds_v5 FAQs