E96-48ds_v5 by Microsoft Azure

E96-48ds_v5 is a EDSv5 family [Local Disk] [Premium Storage capable] [Intel processor] 96 vCPUs server offered by Microsoft Azure with 48 vCPUs, 672 GiB of memory and 3.69 TB of storage. The pricing starts at 0.8352 USD per hour.
48 vCPU
672 GiB Memory
3.69 TB Storage
Spare SCore
46,616
(All-cores)
1,900
(Single-core)

Specifications

Server Metadata

Vendor ID
azure
Server ID
Standard_E96-48ds_v5
Name
E96-48ds_v5
API Reference
Standard_E96-48ds_v5
Description
EDSv5 family [Local Disk] [Premium Storage capable] [Intel processor] 96 vCPUs
Family
EDSv5
Hw Virt
Average Time To Start
27
Status
active
Observed At
2026-11-01T20:55:40.191489

Availability

REGION / IDSPOTONDEMAND
Australia Central 2 (AU) / australiacentral2
0.8352 USD/h -
Sweden Central (SE) / swedencentral
1.3483 USD/h -
East US (US) / eastus
1.4619 USD/h -
Belgium Central (BE) / belgiumcentral
1.5346 USD/h -
Denmark East (DK) / denmarkeast
1.5346 USD/h -
Australia Central (AU) / australiacentral
1.5434 USD/h -
Australia East (AU) / australiaeast
1.5434 USD/h -
Japan East (JP) / japaneast
1.5434 USD/h -
West India (IN) / westindia
1.6349 USD/h -
UAE Central (AE) / uaecentral
2.0295 USD/h -
Norway West (NO) / norwaywest
2.1945 USD/h -
Switzerland West (CH) / switzerlandwest
2.2353 USD/h -
Brazil Southeast (US) / brazilsoutheast
2.6292 USD/h -
North Central US (US) / northcentralus
1.2773 USD/h 6.912 USD/h
East US 2 (US) / eastus2
2.3639 USD/h 6.912 USD/h
West US 2 (US) / westus2
- 6.912 USD/h
West US 3 (US) / westus3
- 6.912 USD/h
Central India (IN) / centralindia
1.3394 USD/h 7.248 USD/h
India South Central (IN) / indiasouthcentral
1.4496 USD/h 7.248 USD/h
Jio India Central (IN) / jioindiacentral
- 7.248 USD/h
Jio India West (IN) / jioindiawest
- 7.248 USD/h
Malaysia West (MY) / malaysiawest
1.3891 USD/h 7.517 USD/h
Indonesia Central (ID) / indonesiacentral
- 7.517 USD/h
Canada East (CA) / canadaeast
1.4015 USD/h 7.584 USD/h
Canada Central (CA) / canadacentral
2.6066 USD/h 7.584 USD/h
Mexico Central (ZA) / mexicocentral
- 7.603 USD/h
Spain Central (ES) / spaincentral
1.4193 USD/h 7.68 USD/h
North Europe (IE) / northeurope
- 7.68 USD/h
West US (US) / westus
1.437 USD/h 7.776 USD/h
Central US (US) / centralus
1.4427 USD/h 7.811 USD/h
Korea South (KR) / koreasouth
1.469 USD/h 7.949 USD/h
Qatar Central (QA) / qatarcentral
1.487 USD/h 8.064 USD/h
Israel Central (IL) / israelcentral
1.4902 USD/h 8.064 USD/h
Italy North (IT) / italynorth
1.4902 USD/h 8.064 USD/h
UK South (GB) / uksouth
1.4269 USD/h 8.112 USD/h
France Central (FR) / francecentral
1.4991 USD/h 8.112 USD/h
South Central US (US) / southcentralus
1.5327 USD/h 8.294 USD/h
West Central US (US) / westcentralus
1.5327 USD/h 8.294 USD/h
Austria East (AT) / austriaeast
1.5346 USD/h 8.304 USD/h
Germany West Central (DE) / germanywestcentral
1.5346 USD/h 8.304 USD/h
Korea Central (KR) / koreacentral
1.5346 USD/h 8.304 USD/h
Poland Central (PL) / polandcentral
1.5346 USD/h 8.304 USD/h
West Europe (NL) / westeurope
1.5346 USD/h 8.304 USD/h
Japan West (JP) / japanwest
1.5434 USD/h 8.352 USD/h
Southeast Asia (SG) / southeastasia
1.5485 USD/h 8.352 USD/h
UK West (GB) / ukwest
1.5457 USD/h 8.364 USD/h
East Asia (HK) / eastasia
1.5523 USD/h 8.4 USD/h
UAE North (AE) / uaenorth
- 8.448 USD/h
New Zealand North (NZ) / newzealandnorth
- 8.77 USD/h
Australia Southeast (AU) / australiasoutheast
1.6606 USD/h 8.986 USD/h
South Africa North (ZA) / southafricanorth
1.6854 USD/h 9.12 USD/h
Switzerland North (CH) / switzerlandnorth
- 9.12 USD/h
Norway East (NO) / norwayeast
1.688 USD/h 9.134 USD/h
Chile Central (CL) / chilecentral
1.7883 USD/h 9.677 USD/h
South India (IN) / southindia
1.8011 USD/h 9.746 USD/h
France South (FR) / francesouth
1.9489 USD/h 10.546 USD/h
Germany North (DE) / germanynorth
1.9949 USD/h 10.795 USD/h
Brazil South (BR) / brazilsouth
2.0225 USD/h 10.944 USD/h
South Africa West (ZA) / southafricawest
2.191 USD/h 11.856 USD/h

Processor

vCPUs
48
Hypervisor
Microsoft Hyper-V
CPU Allocation
Dedicated
CPU Cores
24
CPU Speed
2.8 GHz
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU Family
Xeon
CPU Model
8370C
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
1 MiB
CPU L2 Cache Total
30 MiB
CPU L3 Cache
48 MiB
CPU L3 Cache Total
96 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, vmx, ssse3, fma, cx16, pcid, sse4_1, sse4_2, x2apic, movbe, popcnt, tsc_deadline_timer, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, abm, 3dnowprefetch, tpr_shadow, ept, vpid, ept_ad, 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, vnmi, avx512vbmi, umip, avx512_vbmi2, gfni, vaes, vpclmulqdq, avx512_vnni, avx512_bitalg, avx512_vpopcntdq, la57, rdpid, fsrm, arch_capabilities
Ecpus
24.5
Scalability
51.04

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
    3686 GB
    Storage Type
    ssd
    Storages
    • 3686 GB ssd (temp disk)
    NETWORK
    Inbound Traffic
    0 GB/month
    Outbound Traffic
    0 GB/month
    IPv4
    0

    CPU and System Topology

    Server Description

    A high-capacity memory and storage virtual server designed for demanding database operations and large-scale data processing.

    Memory OptimizedStorage & Database

    Microsoft Azure E96-48ds_v5 is a memory-optimized virtual machine from the EDSv5 family, running on Microsoft Hyper-V. It features 48 dedicated vCPUs via an Intel Xeon 8370C processor at 2.8 GHz, 672.0 GB of RAM, and a 3686 GB local SSD. Designed for memory-intensive applications, it delivers strong performance in large-block memory bandwidth and database operations, though CPU-only LLM inference for large models remains a bottleneck. This server is highly efficient for databases, caching, and high-throughput web serving.

    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) 6,612,953 ops/sec 1,918,343 ops/sec 3.5 50.00% 50.00% +87.00%
    Redis RPS (pipeline=16, SET) 33,217,875 ops/sec 11,704,734 ops/sec 2.8 20.00% 20.00% +23.00%
    PassMark Memory Mark (composite) 2,902 2,416 1.2 10.00% 10.00% +1.80%
    Memory bandwidth (read, 16 MB ~ L3) 269,143 MB/sec 108,466 MB/sec 2.5 10.00% 10.00% +9.60%
    PassMark single-thread CPU 2,436 Mops/s 2,365 Mops/s 1 10.00% 10.00% 0.00%

    Memory Bandwidth

    Compression

    OpenSSL

    Geekbench Single-Core

    Score: 1,708

    Geekbench Multi-Core

    Score: 15,646

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		 50,980
    Compression
    		 730,859
    Encryption
    		 43,847
    Extended Instructions
    		 45,260
    Floating Point Maths
    		 103,273
    Integer Maths
    		 201,995
    Physics
    		 5,824
    Prime Numbers
    		 328
    Single Threaded
    		 2,436
    String Sorting
    		 103,811

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		 2,902
    Database Operations
    		 23,766
    Memory Latency
    		 51
    Memory Read Cached
    		 24,197
    Memory Read Uncached
    		 12,700
    Memory Write
    		 11,408

    Stress-ng Stressors

    CPU integerSingleMultiScaling
    atomic
    		 222.53 240.01 1.08×
    bitops
    		 130,714 3,164,783 24.21×
    branch
    		 137,706 5,226,740 37.96×
    funccall
    		 635.23 15,278 24.05×
    funcret
    		 12,079 416,518 34.48×
    hash
    		 152,173 5,225,967 34.34×
    intmath
    		 61,238 1,791,071 29.25×
    nop
    		 187,642 3,998,956 21.31×
    prime
    		 394,460 10,301,709 26.12×
    rotate
    		 97,991 2,473,743 25.24×
    str
    		 825,069 22,664,766 27.47×
    wcs
    		 615,248 12,954,543 21.06×
    CryptographySingleMultiScaling
    crypt
    		 2,848 63,331 22.23×
    CompressionSingleMultiScaling
    jpeg
    		 726.24 21,829 30.06×
    zlib
    		 56.36 2,025 35.93×
    HypervisorSingleMultiScaling
    clock
    		 31,952 188,856 5.91×
    clone
    		 1,329 23,051 17.34×
    context
    		 5,481 205,780 37.54×
    cpu-sched
    		 58.71 3,070 52.29×
    eventfd
    		 43,315 1,588,476 36.67×
    fault
    		 17,566 170,782 9.72×
    fork
    		 5,364 52,827 9.85×
    getrandom
    		 50,015 1,372,237 27.44×
    hrtimers
    		 138,195 6,384,629 46.20×
    mprotect
    		 723,252 12,928,777 17.88×
    mtx
    		 24,352,059 1,324,170,940 54.38×
    munmap
    		 4,104 54,837 13.36×
    null
    		 868,326 29,599,212 34.09×
    pipe
    		 839,119 30,447,879 36.29×
    priv-instr
    		 822,246 14,945,080 18.18×
    resched
    		 9,776,493 4,069,871 0.42×
    switch
    		 75,031 12,485,100 166.40×
    timermix
    		 267,524 9,170,605 34.28×
    tlb-shootdown
    		 6,235 80,941 12.98×
    tsc
    		 900,278 29,493,600 32.76×
    usersyscall
    		 694,841 14,153,154 20.37×
    vdso
    		 119,952,799 3,017,602,018 25.16×
    vfork
    		 28,129 181,019 6.44×
    x86syscall
    		 10,786,995 394,509,780 36.57×
    yield
    		 49,222,939 51,632,907 1.05×
    zero
    		 4,608,293 132,452,165 28.74×
    CPU float/SIMDSingleMultiScaling
    eigen
    		 14,820 438,464 29.58×
    expmath
    		 2,898 93,476 32.25×
    fma
    		 659,411 25,539,278 38.73×
    fp
    		 192 4,938 25.72×
    fractal
    		 5 175.73 35.15×
    hyperbolic
    		 3,088 107,702 34.88×
    logmath
    		 4,856 168,073 34.61×
    matrix
    		 3,767 128,583 34.14×
    matrix-3d
    		 745.94 8,004 10.73×
    monte-carlo
    		 280.53 9,646 34.39×
    mpfr
    		 1,416 36,210 25.56×
    powmath
    		 1,692 59,907 35.41×
    trig
    		 1,982 59,789 30.17×
    veccmp
    		 20,846 471,533 22.62×
    vecfp
    		 1,531 45,108 29.45×
    vecmath
    		 3,290 112,344 34.14×
    vecshuf
    		 667.22 17,310 25.94×
    vecwide
    		 31,942 882,943 27.64×
    AlgorithmsSingleMultiScaling
    bitonicsort
    		 8 192.39 24.05×
    bsearch
    		 247.41 7,942 32.10×
    bubblesort
    		 1 19.2 19.20×
    fibsearch
    		 290.45 8,489 29.23×
    heapsort
    		 9.4 262.19 27.89×
    hsearch
    		 7,265 194,729 26.81×
    insertionsort
    		 10.4 151.2 14.54×
    malloc
    		 2,625,071 38,951,859 14.84×
    mcontend
    		 65.33 8,281 126.76×
    mergesort
    		 26.8 888 33.13×
    qsort
    		 20.6 521.39 25.31×
    skiplist
    		 5,439 188,586 34.67×
    tree
    		 1.6 38.4 24.00×
    MemorySingleMultiScaling
    cache
    		 248,060 1,193,899 4.81×
    cachehammer
    		 343,861 10,127,109 29.45×
    cacheline
    		 42.76 3,130 73.20×
    l1cache
    		 919.55 9,515 10.35×
    memcpy
    		 408.92 9,603 23.48×
    memrate
    		 1.78 27.34 15.36×
    memthrash
    		 110.59 816.94 7.39×
    ptr-chase
    		 41,209,837 522,263,625 12.67×
    spinmem
    		 411.59 1,870 4.54×
    stream
    		 20.53 14,587 710.50×
    vm
    		 158,433 300,964 1.90×

    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
    E2ds_v5 216 GiB0
    E4-2ds_v5 232 GiB0
    E8-2ds_v5 264 GiB0
    E4ds_v5 432 GiB0
    E8-4ds_v5 464 GiB0
    E16-4ds_v5 4128 GiB0
    E8ds_v5 864 GiB0

    Similar Servers

    INSTANCEVENDORvCPUsMEMORYAccelerators

    E96-48ds_v5 FAQs