m9gd.24xlarge by Amazon Web Services

m9gd.24xlarge is a General purpose [AWS Graviton processors] [Instance store volumes] Gen9 24xlarge server offered by Amazon Web Services with 96 vCPUs, 384 GiB of memory and 5.7 TB of storage. The pricing starts at 0.9557 USD per hour.
96 vCPU
384 GiB Memory
5.7 TB Storage
Spare SCore
191,709
(All-cores)
2,006
(Single-core)

Specifications

Server Metadata

Vendor ID
aws
Name
m9gd.24xlarge
Description
General purpose [AWS Graviton processors] [Instance store volumes] Gen9 24xlarge
Family
m9gd
Hw Virt
Average Time To Start
15
Status
active
Observed At
2026-10-28T23:25:16.782738

Availability

REGION / IDSPOTONDEMAND
Ohio (US) / us-east-2
1.1109 USD/h 6.0336 USD/h
Northern Virgina (US) / us-east-1
1.4925 USD/h 6.0336 USD/h
Oregon (US) / us-west-2
1.9903 USD/h 6.0336 USD/h
(MY) / ap-southeast-5
1.8792 USD/h 6.3994 USD/h
Aragón (ES) / eu-south-2
1.08945 USD/h 6.7286 USD/h
Dublin (IE) / eu-west-1
2.957333 USD/h 6.7286 USD/h
Frankfurt (DE) / eu-central-1
1.956967 USD/h 7.2624 USD/h
Singapore (SG) / ap-southeast-1
2.1226 USD/h 7.5293 USD/h
Sydney (AU) / ap-southeast-2
1.300167 USD/h 7.5821 USD/h
Tokyo (JP) / ap-northeast-1
1.5507 USD/h 7.7952 USD/h

Processor

vCPUs
96
Hypervisor
nitro
CPU Allocation
Dedicated
CPU Cores
96
CPU Speed
3.3 GHz
CPU Architecture
arm64
CPU Manufacturer
AWS
CPU Family
ARMv9
CPU Model
Neoverse-V3
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
2 MiB
CPU L2 Cache Total
192 MiB
CPU L3 Cache
96 MiB
CPU L3 Cache Total
96 MiB
CPU Flags
fp, asimd, evtstrm, aes, pmull, sha1, sha2, crc32, atomics, fphp, asimdhp, cpuid, asimdrdm, jscvt, fcma, lrcpc, dcpop, sha3, asimddp, sha512, sve, asimdfhm, dit, uscat, ilrcpc, flagm, sb, paca, pacg, dcpodp, sve2, sveaes, svepmull, svebitperm, svesha3, flagm2, frint, svei8mm, svebf16, i8mm, bf16, dgh, rng, bti, ecv, afp, wfxt
Ecpus
95.6
Scalability
99.58

System Resources and Accelerators

MEMORY
Memory Amount
384 GiB
Memory Amount Actual
378 GiB
ACCELERATORS
Accelerator Count
0
Accelerator Memory Min
0 MiB
Accelerator Memory Total
0 MiB
Accelerators
    STORAGE
    Storage Size
    5700 GB
    Storage Type
    nvme ssd
    Storages
    • 1900 GB nvme ssd
    • 1900 GB nvme ssd
    • 1900 GB nvme ssd
    NETWORK
    Network Speed Baseline
    50 Gbps
    Network Speed Max
    50 Gbps
    Network Storage Speed Baseline
    36 Gbps
    Network Storage Speed Max
    36 Gbps
    Inbound Traffic
    0 GB/month
    Outbound Traffic
    0 GB/month
    IPv4
    0

    CPU and System Topology

    Server Description

    A high-density ARM64 server featuring dedicated physical cores, massive local NVMe storage, and top-tier multi-core database performance.

    General PurposeStorage & DatabaseCompute Optimized

    Amazon Web Services m9gd.24xlarge is a general-purpose server powered by 96 dedicated physical ARMv9 Neoverse-V3 cores running at 3.3 GHz. It features 384.0 GB of memory, 5700 GB of local NVMe SSD storage, and a 50 Gbps baseline network bandwidth managed by the Nitro hypervisor. Benchmarks indicate top-tier multi-core capabilities, with a stress-ng multi-core score of 191709.22 and exceptional memory read bandwidth of up to 13843.78 GB/sec. The server delivers strong database performance, handling 268.05 million Redis SET operations per second, and excels in software compilation. This resource-dense, non-hyperthreaded hardware configuration offers predictable performance and qualitative cost efficiency for demanding workloads. It is highly suited for large-scale databases, high-throughput web serving, caching layers, and medium-scale machine learning inference.

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    8.7Score
    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) 32,651,775 ops/sec 1,919,289 ops/sec 17 50.00% 50.00% +310.00%
    Redis RPS (pipeline=16, SET) 192,024,950 ops/sec 11,728,058 ops/sec 16 20.00% 20.00% +74.00%
    PassMark Memory Mark (composite) 4,595 2,416 1.9 10.00% 10.00% +6.60%
    Memory bandwidth (read, 16 MB ~ L3) 422,468 MB/sec 108,466 MB/sec 3.8 10.00% 10.00% +14.00%
    PassMark single-thread CPU 2,461 Mops/s 2,365 Mops/s 1 10.00% 10.00% 0.00%

    Memory Bandwidth

    Compression

    OpenSSL

    Geekbench Single-Core

    Score: 2,361

    Geekbench Multi-Core

    Score: 31,014

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		 77,092
    Compression
    		 889,570
    Encryption
    		 28,479
    Extended Instructions
    		 128,857
    Floating Point Maths
    		 731,879
    Integer Maths
    		 585,513
    Physics
    		 20,964
    Prime Numbers
    		 1,110
    Single Threaded
    		 2,461
    String Sorting
    		 491,288

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		 4,595
    Database Operations
    		 57,407
    Memory Latency
    		 29
    Memory Read Cached
    		 24,854
    Memory Read Uncached
    		 24,495
    Memory Write
    		 24,968

    Stress-ng Stressors

    CPU integerSingleMultiScaling
    atomic
    		 175.44 9.26 0.05×
    bitops
    		 165,668 15,783,284 95.27×
    branch
    		 319,821 30,570,688 95.59×
    funccall
    		 2,279 217,740 95.53×
    funcret
    		 40,333 3,745,697 92.87×
    hash
    		 181,154 17,308,823 95.55×
    intmath
    		 39,522 3,783,505 95.73×
    nop
    		 429,498 41,028,678 95.53×
    prime
    		 600,024 57,357,229 95.59×
    rotate
    		 148,594 14,221,176 95.70×
    str
    		 936,523 89,136,772 95.18×
    wcs
    		 281,180 26,740,411 95.10×
    CryptographySingleMultiScaling
    crypt
    		 4,285 409,701 95.60×
    CompressionSingleMultiScaling
    jpeg
    		 899.78 68,545 76.18×
    zlib
    		 54.49 4,396 80.68×
    HypervisorSingleMultiScaling
    clock
    		 46,420 262,642 5.66×
    clone
    		 351.11 11,844 33.73×
    context
    		 3,949 380,044 96.23×
    cpu-sched
    		 71.05 6,106 85.94×
    eventfd
    		 135,410 4,738,275 34.99×
    fault
    		 28,034 388,117 13.84×
    fork
    		 6,058 69,537 11.48×
    getrandom
    		 73,986 7,068,530 95.54×
    hrtimers
    		 1,646,725 9,443,870 5.73×
    mprotect
    		 1,227,734 19,296,364 15.72×
    mtx
    		 11,541,703 5,701,282,642 493.97×
    munmap
    		 4,553 123,687 27.17×
    null
    		 681,691 65,587,714 96.21×
    pipe
    		 1,142,254 50,385,802 44.11×
    priv-instr
    		 1,109,747 11,411,475 10.28×
    resched
    		 1,650,132 17,294,848 10.48×
    switch
    		 220,801 65,124,676 294.95×
    timermix
    		 592,914 31,089,991 52.44×
    tlb-shootdown
    		 16,129 273,360 16.95×
    vdso
    		 64,098,553 6,093,738,302 95.07×
    vfork
    		 50,744 364,827 7.19×
    yield
    		 153,704,186 173,625,693 1.13×
    zero
    		 3,258,028 312,190,132 95.82×
    CPU float/SIMDSingleMultiScaling
    eigen
    		 2,242 214,202 95.54×
    expmath
    		 424.72 40,594 95.58×
    fma
    		 851,804 81,520,165 95.70×
    fp
    		 179.74 16,793 93.43×
    fractal
    		 9.2 865.74 94.10×
    hyperbolic
    		 407.83 38,977 95.57×
    logmath
    		 490.79 46,900 95.56×
    matrix
    		 7,967 760,441 95.45×
    matrix-3d
    		 1,467 17,135 11.68×
    monte-carlo
    		 269.36 25,962 96.38×
    mpfr
    		 1,888 180,130 95.43×
    powmath
    		 281.76 26,915 95.53×
    trig
    		 553 52,858 95.58×
    veccmp
    		 26,308 2,514,931 95.60×
    vecfp
    		 1,860 177,402 95.37×
    vecmath
    		 4,809 459,428 95.53×
    vecshuf
    		 1,026 98,373 95.93×
    vecwide
    		 30,171 2,878,000 95.39×
    AlgorithmsSingleMultiScaling
    bitonicsort
    		 11.2 1,074 95.93×
    bsearch
    		 421.91 40,335 95.60×
    bubblesort
    		 1.4 134.4 96.00×
    fibsearch
    		 413.86 39,528 95.51×
    heapsort
    		 15.4 1,459 94.74×
    hsearch
    		 10,791 1,026,817 95.16×
    insertionsort
    		 7.6 737.4 97.03×
    malloc
    		 4,063,983 186,813,051 45.97×
    mcontend
    		 213.84 239,349 1119.29×
    mergesort
    		 38.6 3,666 94.97×
    qsort
    		 26.8 2,538 94.71×
    skiplist
    		 7,648 731,210 95.61×
    tree
    		 1.8 95.63 53.13×
    MemorySingleMultiScaling
    cache
    		 709,906 2,566,294 3.61×
    cachehammer
    		 2,291,118 35,433,088 15.47×
    cacheline
    		 77.76 5,507 70.82×
    l1cache
    		 390.14 35,944 92.13×
    memcpy
    		 1,014 91,458 90.18×
    memrate
    		 6.7 53.52 7.99×
    memthrash
    		 272.53 533.77 1.96×
    ptr-chase
    		 28,938,616 861,594,167 29.77×
    spinmem
    		 3,320 91,436 27.54×
    stream
    		 54.28 124,480 2293.29×
    vm
    		 202,185 3,151,429 15.59×

    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
    m9gd.medium 14 GiB0
    m9gd.large 28 GiB0
    m9gd.xlarge 416 GiB0
    m9gd.2xlarge 832 GiB0
    m9gd.4xlarge 1664 GiB0
    m9gd.8xlarge 32128 GiB0
    m9gd.12xlarge 48192 GiB0

    Similar Servers

    INSTANCEVENDORvCPUsMEMORYAccelerators

    m9gd.24xlarge FAQs