m9gd.12xlarge by Amazon Web Services

m9gd.12xlarge is a General purpose [AWS Graviton processors] [Instance store volumes] Gen9 12xlarge server offered by Amazon Web Services with 48 vCPUs, 192 GiB of memory and 2.85 TB of storage. The pricing starts at 0.5283 USD per hour.
48 vCPU
192 GiB Memory
2.85 TB Storage
Spare SCore
96,247
(All-cores)
2,010
(Single-core)

Specifications

Server Metadata

Vendor ID
aws
Name
m9gd.12xlarge
Description
General purpose [AWS Graviton processors] [Instance store volumes] Gen9 12xlarge
Family
m9gd
Hw Virt
Average Time To Start
17.5
Status
active
Observed At
2026-10-24T11:37:54.184292

Availability

REGION / IDSPOTONDEMAND
Ohio (US) / us-east-2
0.679767 USD/h 3.0168 USD/h
Northern Virgina (US) / us-east-1
0.868425 USD/h 3.0168 USD/h
Oregon (US) / us-west-2
1.05885 USD/h 3.0168 USD/h
(MY) / ap-southeast-5
0.996133 USD/h 3.1997 USD/h
Aragón (ES) / eu-south-2
0.72625 USD/h 3.3643 USD/h
Dublin (IE) / eu-west-1
1.263233 USD/h 3.3643 USD/h
Frankfurt (DE) / eu-central-1
1.083533 USD/h 3.6312 USD/h
Singapore (SG) / ap-southeast-1
1.187 USD/h 3.7646 USD/h
Sydney (AU) / ap-southeast-2
0.922167 USD/h 3.791 USD/h
Tokyo (JP) / ap-northeast-1
1.2817 USD/h 3.8976 USD/h

Processor

vCPUs
48
Hypervisor
nitro
CPU Allocation
Dedicated
CPU Cores
48
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
3 MiB
CPU L1I Cache
64 KiB
CPU L1I Cache Total
3 MiB
CPU L2 Cache
2 MiB
CPU L2 Cache Total
96 MiB
CPU L3 Cache
48 MiB
CPU L3 Cache Total
48 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
47.9
Scalability
99.79

System Resources and Accelerators

MEMORY
Memory Amount
192 GiB
Memory Amount Actual
189 GiB
ACCELERATORS
Accelerator Count
0
Accelerator Memory Min
0 MiB
Accelerator Memory Total
0 MiB
Accelerators
    STORAGE
    Storage Size
    2850 GB
    Storage Type
    nvme ssd
    Storages
    • 950 GB nvme ssd
    • 950 GB nvme ssd
    • 950 GB nvme ssd
    NETWORK
    Network Speed Baseline
    25 Gbps
    Network Speed Max
    25 Gbps
    Network Storage Speed Baseline
    18 Gbps
    Network Storage Speed Max
    18 Gbps
    Inbound Traffic
    0 GB/month
    Outbound Traffic
    0 GB/month
    IPv4
    0

    CPU and System Topology

    Server Description

    A balanced ARMv9 platform featuring dedicated compute cores, high-speed local NVMe storage, and top-tier database and memory performance.

    General PurposeStorage & Database

    Amazon Web Services m9gd.12xlarge is a general-purpose server powered by a 3.3 GHz AWS Graviton ARMv9 Neoverse-V3 processor with 48 dedicated physical cores and 192.0 GB of system memory. It features 2850 GB of local NVMe SSD storage and a baseline network bandwidth of 25 Gbps. Benchmark results show top-tier performance in software compilation, database operations, Redis SET operations, and static web serving. While single-core and multi-core Geekbench scores are in the top 10%, text generation for small language models is a relative bottleneck. This instance provides a cost-efficient hardware profile for workloads requiring high-speed local storage and balanced compute resources, making it suitable for database hosting, web serving, and CI/CD compilation pipelines.

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    5.2Score
    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) 16,911,029 ops/sec 1,919,289 ops/sec 8.9 50.00% 50.00% +200.00%
    Redis RPS (pipeline=16, SET) 98,690,026 ops/sec 11,728,058 ops/sec 8.2 20.00% 20.00% +52.00%
    PassMark Memory Mark (composite) 4,658 2,416 2 10.00% 10.00% +7.20%
    Memory bandwidth (read, 16 MB ~ L3) 218,782 MB/sec 108,466 MB/sec 2 10.00% 10.00% +7.20%
    PassMark single-thread CPU 2,468 Mops/s 2,365 Mops/s 1 10.00% 10.00% 0.00%

    Memory Bandwidth

    Compression

    OpenSSL

    Geekbench Single-Core

    Score: 2,396

    Geekbench Multi-Core

    Score: 28,199

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		 49,413
    Compression
    		 448,163
    Encryption
    		 14,327
    Extended Instructions
    		 64,637
    Floating Point Maths
    		 367,360
    Integer Maths
    		 293,877
    Physics
    		 10,681
    Prime Numbers
    		 589
    Single Threaded
    		 2,468
    String Sorting
    		 247,364

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		 4,658
    Database Operations
    		 46,542
    Memory Latency
    		 27
    Memory Read Cached
    		 24,911
    Memory Read Uncached
    		 24,689
    Memory Write
    		 24,941

    Stress-ng Stressors

    CPU integerSingleMultiScaling
    atomic
    		 185.06 19.4 0.10×
    bitops
    		 165,872 7,929,997 47.81×
    branch
    		 320,495 15,346,789 47.88×
    funccall
    		 2,284 109,321 47.87×
    funcret
    		 40,250 1,913,548 47.54×
    hash
    		 181,651 8,691,959 47.85×
    intmath
    		 39,674 1,899,395 47.87×
    nop
    		 430,208 20,592,516 47.87×
    prime
    		 601,333 28,784,906 47.87×
    rotate
    		 149,182 7,142,714 47.88×
    str
    		 936,042 44,921,593 47.99×
    wcs
    		 284,650 13,359,793 46.93×
    CryptographySingleMultiScaling
    crypt
    		 4,367 209,015 47.86×
    CompressionSingleMultiScaling
    jpeg
    		 925.03 43,505 47.03×
    zlib
    		 45.29 2,247 49.62×
    HypervisorSingleMultiScaling
    clock
    		 51,434 500,928 9.74×
    clone
    		 397.97 8,299 20.85×
    context
    		 3,955 192,428 48.65×
    cpu-sched
    		 70.08 3,092 44.12×
    eventfd
    		 156,840 3,365,584 21.46×
    fault
    		 29,782 483,124 16.22×
    fork
    		 7,468 78,110 10.46×
    getrandom
    		 74,119 3,548,278 47.87×
    hrtimers
    		 505,545 5,750,678 11.38×
    mprotect
    		 1,424,861 23,763,917 16.68×
    mtx
    		 9,702,466 2,724,877,693 280.84×
    munmap
    		 5,363 130,166 24.27×
    null
    		 683,356 33,203,241 48.59×
    pipe
    		 1,255,766 33,668,704 26.81×
    priv-instr
    		 1,120,320 14,300,122 12.76×
    resched
    		 3,581,933 9,670,544 2.70×
    switch
    		 276,352 33,680,171 121.87×
    timermix
    		 594,603 20,591,996 34.63×
    tlb-shootdown
    		 19,785 275,175 13.91×
    vdso
    		 62,231,517 3,035,292,124 48.77×
    vfork
    		 63,661 496,111 7.79×
    yield
    		 79,374,913 93,687,565 1.18×
    zero
    		 3,263,127 157,344,993 48.22×
    CPU float/SIMDSingleMultiScaling
    eigen
    		 2,248 107,523 47.84×
    expmath
    		 425.59 20,376 47.88×
    fma
    		 853,531 40,907,990 47.93×
    fp
    		 178.98 8,416 47.02×
    fractal
    		 9.2 435.27 47.31×
    hyperbolic
    		 408.79 19,560 47.85×
    logmath
    		 491.73 23,543 47.88×
    matrix
    		 8,001 382,816 47.84×
    matrix-3d
    		 1,493 8,858 5.93×
    monte-carlo
    		 269.64 13,060 48.44×
    mpfr
    		 1,890 90,423 47.83×
    powmath
    		 282.35 13,515 47.87×
    trig
    		 553.97 26,525 47.88×
    veccmp
    		 26,377 1,262,059 47.85×
    vecfp
    		 1,860 88,499 47.57×
    vecmath
    		 4,826 230,657 47.79×
    vecshuf
    		 1,033 49,331 47.73×
    vecwide
    		 30,217 1,446,909 47.88×
    AlgorithmsSingleMultiScaling
    bitonicsort
    		 11.4 537.59 47.16×
    bsearch
    		 422.37 20,265 47.98×
    bubblesort
    		 1.4 67.2 48.00×
    fibsearch
    		 415.93 19,849 47.72×
    heapsort
    		 15.4 738.39 47.95×
    hsearch
    		 10,870 520,734 47.91×
    insertionsort
    		 8 369.2 46.15×
    malloc
    		 4,635,037 129,023,098 27.84×
    mcontend
    		 308.26 115,593 374.98×
    mergesort
    		 38.6 1,839 47.63×
    qsort
    		 26.8 1,285 47.93×
    skiplist
    		 7,678 367,282 47.84×
    tree
    		 2.2 48.11 21.87×
    MemorySingleMultiScaling
    cache
    		 698,322 1,180,533 1.69×
    cachehammer
    		 2,406,280 53,721,432 22.33×
    cacheline
    		 51.98 6,295 121.11×
    l1cache
    		 390.08 18,542 47.53×
    memcpy
    		 1,016 45,942 45.20×
    memrate
    		 6.72 31.04 4.62×
    memthrash
    		 338.62 682.42 2.02×
    ptr-chase
    		 34,305,279 462,588,518 13.48×
    spinmem
    		 4,486 57,747 12.87×
    stream
    		 116.56 67,193 576.47×
    vm
    		 200,986 3,770,110 18.76×

    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.16xlarge 64256 GiB0

    Similar Servers

    INSTANCEVENDORvCPUsMEMORYAccelerators

    m9gd.12xlarge FAQs