m9gd.metal-48xl by Amazon Web Services

m9gd.metal-48xl is a General purpose [AWS Graviton processors] [Instance store volumes] Gen9 metal-48xl server offered by Amazon Web Services with 192 vCPUs, 768 GiB of memory and 11.4 TB of storage. The pricing starts at 1.465 USD per hour.
192 vCPU
768 GiB Memory
11.4 TB Storage
Spare SCore
380,524
(All-cores)
2,014
(Single-core)

Specifications

Server Metadata

Vendor ID
aws
Name
m9gd.metal-48xl
Description
General purpose [AWS Graviton processors] [Instance store volumes] Gen9 metal-48xl
Family
m9gd
Hw Virt
Average Time To Start
23
Status
active
Observed At
2026-10-28T23:25:16.784603

Availability

REGION / IDSPOTONDEMAND
Ohio (US) / us-east-2
1.799367 USD/h 12.0672 USD/h
Northern Virgina (US) / us-east-1
2.96705 USD/h 12.0672 USD/h
Oregon (US) / us-west-2
3.410375 USD/h 12.0672 USD/h
(MY) / ap-southeast-5
3.483833 USD/h 12.7987 USD/h
Aragón (ES) / eu-south-2
2.6722 USD/h 13.4573 USD/h
Dublin (IE) / eu-west-1
6.230733 USD/h 13.4573 USD/h
Frankfurt (DE) / eu-central-1
2.728767 USD/h 14.5248 USD/h
Singapore (SG) / ap-southeast-1
4.715 USD/h 15.0586 USD/h
Sydney (AU) / ap-southeast-2
2.936433 USD/h 15.1642 USD/h
Tokyo (JP) / ap-northeast-1
2.7764 USD/h 15.5904 USD/h

Processor

vCPUs
192
CPU Allocation
Dedicated
CPU Cores
192
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
12 MiB
CPU L1I Cache
64 KiB
CPU L1I Cache Total
12 MiB
CPU L2 Cache
2 MiB
CPU L2 Cache Total
384 MiB
CPU L3 Cache
96 MiB
CPU L3 Cache Total
192 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
189
Scalability
98.44

System Resources and Accelerators

MEMORY
Memory Amount
768 GiB
Memory Amount Actual
768 GiB
Memory Generation
DDR5
ACCELERATORS
Accelerator Count
0
Accelerator Memory Min
0 MiB
Accelerator Memory Total
0 MiB
Accelerators
    STORAGE
    Storage Size
    11400 GB
    Storage Type
    nvme ssd
    Storages
    • 3800 GB nvme ssd
    • 3800 GB nvme ssd
    • 3800 GB nvme ssd
    NETWORK
    Network Speed Baseline
    100 Gbps
    Network Speed Max
    100 Gbps
    Network Storage Speed Baseline
    72 Gbps
    Network Storage Speed Max
    72 Gbps
    Inbound Traffic
    0 GB/month
    Outbound Traffic
    0 GB/month
    IPv4
    0

    CPU and System Topology

    Server Description

    A high-density bare-metal ARM platform featuring massive local NVMe storage and exceptional multi-core throughput for database and enterprise workloads.

    General PurposeStorage & Database

    Amazon Web Services m9gd.metal-48xl is a bare-metal, general-purpose server featuring 192 dedicated physical cores of AWS Graviton ARMv9 architecture running at 3.3 GHz. It is equipped with 768.0 GB of DDR5 memory, 11,400 GB of local NVMe SSD storage, and a 100 Gbps baseline network interface. Lacking a GPU, the server relies on its high-density ARM architecture to deliver top-tier multi-core performance, as evidenced by a PassMark CPU score of 107,915.98 and a multi-threaded bzip3 decompression rate of 1,953.03 GB/sec. It excels in memory bandwidth and database operations, achieving 520.51 million operations per second in Redis SET benchmarks. This configuration is highly suitable for large-scale databases, high-throughput web serving, and parallel data processing, though it exhibits lower performance in single-threaded tasks and LLM text generation.

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    15Score
    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) 64,925,203 ops/sec 1,919,289 ops/sec 34 50.00% 50.00% +480.00%
    Redis RPS (pipeline=16, SET) 384,360,491 ops/sec 11,728,058 ops/sec 32 20.00% 20.00% +100.00%
    PassMark Memory Mark (composite) 4,596 2,416 1.9 10.00% 10.00% +6.60%
    Memory bandwidth (read, 16 MB ~ L3) 845,905 MB/sec 108,466 MB/sec 7.7 10.00% 10.00% +23.00%
    PassMark single-thread CPU 2,474 Mops/s 2,365 Mops/s 1 10.00% 10.00% 0.00%

    Memory Bandwidth

    Compression

    OpenSSL

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		 107,916
    Compression
    		 1,787,085
    Encryption
    		 56,832
    Extended Instructions
    		 244,219
    Floating Point Maths
    		 1,464,559
    Integer Maths
    		 1,168,329
    Physics
    		 42,807
    Prime Numbers
    		 2,297
    Single Threaded
    		 2,474
    String Sorting
    		 930,151

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		 4,596
    Database Operations
    		 62,497
    Memory Latency
    		 30
    Memory Read Cached
    		 25,008
    Memory Read Uncached
    		 24,689
    Memory Write
    		 25,060

    Stress-ng Stressors

    CPU integerSingleMultiScaling
    atomic
    		 137.78 - -
    bitops
    		 165,707 12,258,520 73.98×
    branch
    		 321,291 24,064,670 74.90×
    funccall
    		 2,289 430,259 187.95×
    funcret
    		 40,314 7,227,604 179.28×
    hash
    		 181,948 34,677,191 190.59×
    intmath
    		 39,660 7,498,514 189.07×
    nop
    		 431,189 81,144,080 188.19×
    prime
    		 602,879 114,670,196 190.20×
    rotate
    		 149,624 28,418,915 189.94×
    str
    		 940,381 177,602,526 188.86×
    wcs
    		 284,817 52,328,062 183.73×
    CryptographySingleMultiScaling
    crypt
    		 4,377 781,949 178.67×
    CompressionSingleMultiScaling
    jpeg
    		 921.17 113,351 123.05×
    zlib
    		 53.95 8,923 165.40×
    HypervisorSingleMultiScaling
    clock
    		 45,876 141,141 3.08×
    clone
    		 509.75 22,286 43.72×
    context
    		 3,975 763,604 192.12×
    cpu-sched
    		 67.23 11,870 176.56×
    eventfd
    		 147,116 5,222,630 35.50×
    fault
    		 29,582 290,855 9.83×
    fork
    		 5,490 74,054 13.49×
    getrandom
    		 74,060 14,110,699 190.53×
    hrtimers
    		 130,415 6,287,744 48.21×
    mprotect
    		 917,391 15,406,602 16.79×
    mtx
    		 11,203,633 10,328,667,124 921.90×
    munmap
    		 4,011 106,606 26.58×
    null
    		 682,802 129,999,772 190.39×
    pipe
    		 974,089 86,083,698 88.37×
    priv-instr
    		 1,101,265 7,342,717 6.67×
    resched
    		 1,057,241 23,244,957 21.99×
    switch
    		 291,582 62,470,816 214.25×
    timermix
    		 975,542 179,437,491 183.94×
    tlb-shootdown
    		 14,118 226,982 16.08×
    vdso
    		 64,339,507 11,783,299,865 183.14×
    vfork
    		 42,285 277,904 6.57×
    yield
    		 365,628,984 365,663,320 1.00×
    zero
    		 3,272,303 622,339,946 190.18×
    CPU float/SIMDSingleMultiScaling
    eigen
    		 2,253 417,260 185.16×
    expmath
    		 426.49 79,794 187.09×
    fma
    		 857,117 161,840,504 188.82×
    fp
    		 177.6 33,546 188.89×
    fractal
    		 9.2 1,701 184.88×
    hyperbolic
    		 409.59 77,767 189.87×
    logmath
    		 492.74 93,600 189.96×
    matrix
    		 8,021 1,521,296 189.67×
    matrix-3d
    		 1,493 34,538 23.13×
    monte-carlo
    		 269.73 50,606 187.62×
    mpfr
    		 1,897 358,657 189.10×
    powmath
    		 282.95 53,856 190.34×
    trig
    		 555.23 105,058 189.21×
    veccmp
    		 26,423 4,980,549 188.49×
    vecfp
    		 1,865 351,963 188.75×
    vecmath
    		 4,829 916,835 189.85×
    vecshuf
    		 1,030 196,191 190.43×
    vecwide
    		 30,287 6,139,384 202.71×
    AlgorithmsSingleMultiScaling
    bitonicsort
    		 5.6 368.35 65.78×
    bsearch
    		 424.33 65,259 153.79×
    bubblesort
    		 1.4 268.8 192.00×
    fibsearch
    		 413.45 77,683 187.89×
    heapsort
    		 15.4 2,912 189.07×
    hsearch
    		 10,863 2,073,466 190.87×
    insertionsort
    		 8 1,480 185.05×
    malloc
    		 4,375,957 316,068,289 72.23×
    mcontend
    		 160.1 497,302 3106.20×
    mergesort
    		 38.8 7,303 188.22×
    qsort
    		 26.8 5,114 190.84×
    skiplist
    		 7,680 1,440,937 187.62×
    tree
    		 2 191.54 95.77×
    MemorySingleMultiScaling
    cache
    		 604,782 7,633,048 12.62×
    cachehammer
    		 2,168,007 21,294,735 9.82×
    cacheline
    		 56.17 4,211 74.97×
    l1cache
    		 390.25 58,224 149.20×
    memcpy
    		 1,017 174,408 171.47×
    memrate
    		 6.92 94.4 13.64×
    memthrash
    		 493.95 367.47 0.74×
    ptr-chase
    		 29,604,545 1,738,369,859 58.72×
    spinmem
    		 3,069 180,807 58.92×
    stream
    		 27.38 234,032 8547.56×
    vm
    		 203,050 5,211,171 25.66×

    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.metal-48xl FAQs