E96pds_v6 by Microsoft Azure

E96pds_v6 is a Epdsv6 family [ARM processor] [Local Disk] [Premium Storage capable] 96 vCPUs server offered by Microsoft Azure with 96 vCPUs, 672 GiB of memory and 5.67 TB of storage. The pricing starts at 0.6812 USD per hour.
96 vCPU
672 GiB Memory
5.67 TB Storage
Spare SCore
198,369
(All-cores)
2,073
(Single-core)

Specifications

Server Metadata

Vendor ID
azure
Server ID
Standard_E96pds_v6
Name
E96pds_v6
API Reference
Standard_E96pds_v6
Description
Epdsv6 family [ARM processor] [Local Disk] [Premium Storage capable] 96 vCPUs
Family
Epdsv6
Hw Virt
Average Time To Start
138
Status
active
Observed At
2026-10-27T15:00:06.175744

Availability

REGION / IDSPOTONDEMAND
Central India (IN) / centralindia
0.6812 USD/h 3.686 USD/h
East US 2 (US) / eastus2
1.0378 USD/h 5.616 USD/h
North Central US (US) / northcentralus
1.0378 USD/h 5.616 USD/h
West US 2 (US) / westus2
1.0378 USD/h 5.616 USD/h
West US 3 (US) / westus3
1.0378 USD/h 5.616 USD/h
East US (US) / eastus
2.4884 USD/h 5.616 USD/h
Sweden Central (SE) / swedencentral
1.1053 USD/h 5.981 USD/h
Mexico Central (ZA) / mexicocentral
1.1408 USD/h 6.173 USD/h
Canada Central (CA) / canadacentral
1.1532 USD/h 6.24 USD/h
Canada East (CA) / canadaeast
1.1532 USD/h 6.24 USD/h
Spain Central (ES) / spaincentral
1.1585 USD/h 6.269 USD/h
North Europe (IE) / northeurope
1.7039 USD/h 6.269 USD/h
Central US (US) / centralus
1.2299 USD/h 6.346 USD/h
UK South (GB) / uksouth
1.0371 USD/h 6.47 USD/h
Italy North (IT) / italynorth
1.2064 USD/h 6.528 USD/h
France Central (FR) / francecentral
1.2117 USD/h 6.557 USD/h
West US (US) / westus
1.2117 USD/h 6.557 USD/h
Indonesia Central (ID) / indonesiacentral
1.2143 USD/h 6.571 USD/h
Germany West Central (DE) / germanywestcentral
1.24 USD/h 6.71 USD/h
Poland Central (PL) / polandcentral
1.24 USD/h 6.71 USD/h
West Europe (NL) / westeurope
1.3158 USD/h 6.71 USD/h
South Central US (US) / southcentralus
1.2454 USD/h 6.739 USD/h
West Central US (US) / westcentralus
1.2454 USD/h 6.739 USD/h
UK West (GB) / ukwest
1.2542 USD/h 6.787 USD/h
UAE North (AE) / uaenorth
1.2666 USD/h 6.854 USD/h
Australia East (AU) / australiaeast
1.2932 USD/h 6.998 USD/h
Southeast Asia (SG) / southeastasia
1.2932 USD/h 6.998 USD/h
Japan East (JP) / japaneast
1.3341 USD/h 7.219 USD/h
Japan West (JP) / japanwest
1.3359 USD/h 7.229 USD/h
Australia Southeast (AU) / australiasoutheast
1.3483 USD/h 7.296 USD/h
Austria East (AT) / austriaeast
1.4602 USD/h 7.301 USD/h
Belgium Central (BE) / belgiumcentral
1.4602 USD/h 7.301 USD/h
Denmark East (DK) / denmarkeast
1.4602 USD/h 7.301 USD/h
Korea Central (KR) / koreacentral
1.3518 USD/h 7.315 USD/h
New Zealand North (NZ) / newzealandnorth
1.3572 USD/h 7.344 USD/h
East Asia (HK) / eastasia
1.4281 USD/h 7.728 USD/h
South Africa North (ZA) / southafricanorth
1.4547 USD/h 7.872 USD/h
Norway East (NO) / norwayeast
1.4832 USD/h 8.026 USD/h
South India (IN) / southindia
1.696 USD/h 8.48 USD/h
Switzerland North (CH) / switzerlandnorth
1.6179 USD/h 8.755 USD/h
Brazil South (BR) / brazilsouth
1.648 USD/h 8.918 USD/h
South Africa West (ZA) / southafricawest
1.8888 USD/h 10.221 USD/h
Switzerland West (CH) / switzerlandwest
1.9304 USD/h 10.446 USD/h
UAE Central (AE) / uaecentral
2.0892 USD/h 10.446 USD/h

Processor

vCPUs
96
Hypervisor
Microsoft Hyper-V
CPU Allocation
Dedicated
CPU Cores
96
CPU Speed
3.4 GHz
CPU Architecture
arm64
CPU Manufacturer
Microsoft
CPU Family
ARM
CPU Model
Neoverse-N2
CPU L1D Cache
64 KiB
CPU L1D Cache Total
6 MiB
CPU L1I Cache
64 KiB
CPU L1I Cache Total
6 MiB
CPU L2 Cache
1 MiB
CPU L2 Cache Total
96 MiB
CPU L3 Cache
128 MiB
CPU L3 Cache Total
128 MiB
CPU Flags
fp, asimd, evtstrm, aes, pmull, sha1, sha2, crc32, atomics, fphp, asimdhp, cpuid, asimdrdm, jscvt, fcma, lrcpc, dcpop, sha3, sm3, sm4, asimddp, sha512, sve, asimdfhm, uscat, ilrcpc, flagm, ssbs, sb, paca, pacg, dcpodp, sve2, sveaes, svebitperm, svesha3, svesm4, flagm2, frint, svei8mm, svebf16, i8mm, bf16
Ecpus
95.7
Scalability
99.69

System Resources and Accelerators

MEMORY
Memory Amount
672 GiB
Memory Amount Actual
672 GiB
ACCELERATORS
Accelerator Count
0
Accelerator Memory Min
0 MiB
Accelerator Memory Total
0 MiB
Accelerators
    STORAGE
    Storage Size
    5669 GB
    Storage Type
    nvme ssd
    Storages
    • 5669 GB nvme ssd
    NETWORK
    Inbound Traffic
    0 GB/month
    Outbound Traffic
    0 GB/month
    IPv4
    0

    CPU and System Topology

    Server Description

    A 96-core ARM Neoverse-N2 server equipped with 672 GB memory and bundled local NVMe storage for memory-intensive data workloads.

    Memory OptimizedStorage & DatabaseCompute Optimized

    Microsoft Azure E96pds_v6 is a memory-optimized cloud server built on the ARM-based Microsoft Neoverse-N2 processor operating at 3.4 GHz. It features 96 dedicated physical cores and vCPUs, 672 GB of system memory, and 5,669 GB of bundled NVMe SSD local storage. Benchmark results show top-tier multi-threaded CPU performance and exceptional memory bandwidth exceeding 465 GB/s read speeds, alongside high throughput in database operations and Redis SET workloads reaching 170.49 million operations per second. Single-core performance remains in the average benchmark tier, highlighting a tradeoff optimized for parallel scaling rather than single-threaded clock speed. With high memory density and local high-speed storage, the instance delivers effective resource utilization for memory-bound applications. The E96pds_v6 is well suited for high-density database clusters, distributed caching, static web serving, and multi-threaded data processing workloads.

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    5.4Score
    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) 17,061,297 ops/sec 1,918,343 ops/sec 8.9 50.00% 50.00% +200.00%
    Redis RPS (pipeline=16, SET) 104,494,694 ops/sec 11,704,734 ops/sec 8.3 20.00% 20.00% +53.00%
    PassMark Memory Mark (composite) 3,448 2,416 1.4 10.00% 10.00% +3.40%
    Memory bandwidth (read, 16 MB ~ L3) 509,277 MB/sec 108,420 MB/sec 4.6 10.00% 10.00% +16.00%
    PassMark single-thread CPU 1,885 Mops/s 2,365 Mops/s 0.79 10.00% 10.00% -2.30%

    Memory Bandwidth

    Compression

    OpenSSL

    Geekbench Single-Core

    Score: 1,689

    Geekbench Multi-Core

    Score: 23,825

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		 56,681
    Compression
    		 657,254
    Encryption
    		 18,139
    Extended Instructions
    		 85,661
    Floating Point Maths
    		 507,335
    Integer Maths
    		 603,773
    Physics
    		 18,268
    Prime Numbers
    		 632
    Single Threaded
    		 1,885
    String Sorting
    		 477,752

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		 3,448
    Database Operations
    		 37,969
    Memory Latency
    		 52
    Memory Read Cached
    		 25,608
    Memory Read Uncached
    		 19,752
    Memory Write
    		 25,739

    Stress-ng Stressors

    CPU integerSingleMultiScaling
    atomic
    		 205.56 8.58 0.04×
    bitops
    		 151,379 14,468,088 95.58×
    branch
    		 256,633 24,544,218 95.64×
    funccall
    		 1,908 182,401 95.60×
    funcret
    		 34,047 3,160,239 92.82×
    hash
    		 177,686 16,990,261 95.62×
    intmath
    		 38,981 3,742,040 96.00×
    nop
    		 456,763 43,761,948 95.81×
    prime
    		 488,761 46,785,646 95.72×
    rotate
    		 130,703 12,512,545 95.73×
    str
    		 691,988 66,553,890 96.18×
    wcs
    		 217,461 20,838,683 95.83×
    CryptographySingleMultiScaling
    crypt
    		 3,820 367,054 96.09×
    CompressionSingleMultiScaling
    jpeg
    		 688.92 65,967 95.75×
    zlib
    		 35.55 4,393 123.56×
    HypervisorSingleMultiScaling
    clock
    		 26,888 265,614 9.88×
    clone
    		 296.84 9,466 31.89×
    context
    		 2,733 263,731 96.51×
    cpu-sched
    		 72.72 5,762 79.24×
    eventfd
    		 56,805 2,258,507 39.76×
    fault
    		 22,949 358,173 15.61×
    fork
    		 3,739 73,689 19.71×
    getrandom
    		 60,108 5,739,134 95.48×
    hrtimers
    		 346,807 10,014,965 28.88×
    mprotect
    		 1,095,099 19,663,168 17.96×
    mtx
    		 10,600,251 4,259,207,467 401.80×
    munmap
    		 2,836 81,063 28.59×
    null
    		 482,033 46,408,466 96.28×
    pipe
    		 908,418 42,830,311 47.15×
    priv-instr
    		 845,931 10,496,628 12.41×
    resched
    		 2,041,767 11,150,525 5.46×
    switch
    		 73,110 47,613,974 651.26×
    timermix
    		 622,047 58,778,641 94.49×
    tlb-shootdown
    		 15,657 239,138 15.27×
    vdso
    		 46,156,538 4,455,832,846 96.54×
    vfork
    		 26,289 272,583 10.37×
    yield
    		 125,116,916 132,804,033 1.06×
    zero
    		 2,819,502 264,377,588 93.77×
    CPU float/SIMDSingleMultiScaling
    eigen
    		 2,128 203,495 95.62×
    expmath
    		 393.46 37,630 95.64×
    fma
    		 769,035 73,504,532 95.58×
    fp
    		 160.8 15,415 95.87×
    fractal
    		 7.8 738.18 94.64×
    hyperbolic
    		 387.6 37,060 95.61×
    logmath
    		 465.2 44,536 95.74×
    matrix
    		 6,603 632,378 95.77×
    matrix-3d
    		 1,057 21,375 20.22×
    monte-carlo
    		 214.28 19,862 92.69×
    mpfr
    		 1,446 138,385 95.68×
    powmath
    		 272.17 26,057 95.74×
    trig
    		 520.69 49,843 95.73×
    veccmp
    		 19,414 1,856,389 95.62×
    vecfp
    		 913.88 87,385 95.62×
    vecmath
    		 4,872 466,375 95.72×
    vecshuf
    		 996.4 97,502 97.85×
    vecwide
    		 27,518 2,631,813 95.64×
    AlgorithmsSingleMultiScaling
    bitonicsort
    		 9.4 892.98 95.00×
    bsearch
    		 381.89 36,536 95.67×
    bubblesort
    		 1.2 115.2 96.00×
    fibsearch
    		 432.69 41,354 95.58×
    heapsort
    		 13 1,266 97.38×
    hsearch
    		 6,996 669,737 95.73×
    insertionsort
    		 7.8 730.6 93.67×
    malloc
    		 3,068,385 169,459,592 55.23×
    mcontend
    		 217.11 229,900 1058.91×
    mergesort
    		 37.6 3,586 95.37×
    qsort
    		 24.4 2,341 95.95×
    skiplist
    		 7,028 674,156 95.92×
    tree
    		 1.4 72.9 52.07×
    MemorySingleMultiScaling
    cache
    		 434,148 5,231,990 12.05×
    cachehammer
    		 1,637,265 22,655,910 13.84×
    cacheline
    		 76.76 5,508 71.76×
    l1cache
    		 303.1 27,794 91.70×
    memcpy
    		 724.19 69,273 95.66×
    memrate
    		 5.32 71.65 13.47×
    memthrash
    		 239.01 713.94 2.99×
    ptr-chase
    		 25,168,442 746,760,571 29.67×
    spinmem
    		 3,975 95,041 23.91×
    stream
    		 12.61 15,221 1207.04×
    vm
    		 139,851 7,001,353 50.06×

    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
    E2pds_v6 216 GiB0
    E4pds_v6 432 GiB0
    E8pds_v6 864 GiB0
    E16pds_v6 16128 GiB0
    E32pds_v6 32256 GiB0
    E48pds_v6 48384 GiB0
    E64pds_v6 64512 GiB0

    Similar Servers

    INSTANCEVENDORvCPUsMEMORYAccelerators

    E96pds_v6 FAQs