m8g.metal-48xl by Amazon Web Services

m8g.metal-48xl is a General purpose [AWS Graviton processors] Gen8 metal-48xl server offered by Amazon Web Services with 192 vCPUs, 768 GiB of memory and 0 GB of storage. The pricing starts at 0.6159 USD per hour.
192 vCPU
768 GiB Memory
Spare SCore
304,251
(All-cores)
1,649
(Single-core)

Specifications

Server Metadata

Vendor ID
aws
Name
m8g.metal-48xl
Description
General purpose [AWS Graviton processors] Gen8 metal-48xl
Family
m8g
Hw Virt
Average Time To Start
14
Status
active
Observed At
2026-10-27T06:58:17.783332

Availability

REGION / IDSPOTONDEMAND
Hyderabad (IN) / ap-south-2
0.624467 USD/h 6.1594 USD/h
Mumbai (IN) / ap-south-1
1.5267 USD/h 6.1594 USD/h
Ohio (US) / us-east-2
1.784333 USD/h 8.617 USD/h
Northern Virgina (US) / us-east-1
2.7423 USD/h 8.617 USD/h
Oregon (US) / us-west-2
3.3531 USD/h 8.617 USD/h
(MX) / mx-central-1
1.364567 USD/h 9.047 USD/h
(MY) / ap-southeast-5
1.666367 USD/h 9.1565 USD/h
(TH) / ap-southeast-7
1.5487 USD/h 9.1565 USD/h
Stockholm (SE) / eu-north-1
0.9156 USD/h 9.1565 USD/h
Aragón (ES) / eu-south-2
0.9598 USD/h 9.5981 USD/h
Dublin (IE) / eu-west-1
2.777933 USD/h 9.6038 USD/h
Quebec (CA) / ca-central-1
1.2661 USD/h 9.6058 USD/h
Calgary (CA) / ca-west-1
2.36395 USD/h 9.6058 USD/h
London (GB) / eu-west-2
1.168625 USD/h 9.9648 USD/h
Taipei (TW) / ap-east-2
2.132067 USD/h 10.0166 USD/h
Milan (IT) / eu-south-1
1.0053 USD/h 10.0531 USD/h
Paris (FR) / eu-west-3
1.411133 USD/h 10.0531 USD/h
California (US) / us-west-1
1.2109 USD/h 10.0531 USD/h
Frankfurt (DE) / eu-central-1
2.005567 USD/h 10.3238 USD/h
United Arab Emirates / me-central-1
4.3246 USD/h 10.5638 USD/h
Seoul (KR) / ap-northeast-2
1.140875 USD/h 10.5926 USD/h
Sydney (AU) / ap-southeast-2
1.320833 USD/h 10.7712 USD/h
Jakarta (ID) / ap-southeast-3
1.0771 USD/h 10.7712 USD/h
Melbourne (AU) / ap-southeast-4
1.533633 USD/h 10.7712 USD/h
Singapore (SG) / ap-southeast-1
2.604933 USD/h 10.7712 USD/h
Osaka (JP) / ap-northeast-3
1.463133 USD/h 11.1302 USD/h
Tokyo (JP) / ap-northeast-1
1.741433 USD/h 11.1302 USD/h
Zurich (CH) / eu-central-2
1.831433 USD/h 11.3549 USD/h
Cape Town (ZA) / af-south-1
3.331267 USD/h 11.401 USD/h
Hong Kong (HK) / ap-east-1
1.1848 USD/h 11.8483 USD/h
Sao Paulo (BR) / sa-east-1
1.3734 USD/h 13.7338 USD/h

Processor

vCPUs
192
CPU Allocation
Dedicated
CPU Cores
192
CPU Speed
2.7 GHz
CPU Architecture
arm64
CPU Manufacturer
AWS
CPU Family
ARMv9
CPU Model
Neoverse-V2
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
36 MiB
CPU L3 Cache Total
72 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, ssbs, sb, paca, pacg, dcpodp, sve2, sveaes, svepmull, svebitperm, svesha3, flagm2, frint, svei8mm, svebf16, i8mm, bf16, dgh, rng, bti
Ecpus
184.6
Scalability
96.15

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
    0 GB
    Storages
      NETWORK
      Network Speed Baseline
      50 Gbps
      Network Speed Max
      50 Gbps
      Network Storage Speed Baseline
      40 Gbps
      Network Storage Speed Max
      40 Gbps
      Inbound Traffic
      0 GB/month
      Outbound Traffic
      0 GB/month
      IPv4
      0

      CPU and System Topology

      Server Description

      A high-density bare-metal ARM64 server delivering exceptional multi-core performance and DDR5 memory bandwidth for parallelized enterprise workloads.

      General PurposeCompute Optimized

      Amazon Web Services m8g.metal-48xl is a bare-metal, general-purpose server featuring 192 dedicated physical ARMv9 cores powered by AWS Graviton Neoverse-V2 processors running at 2.7 GHz. It is equipped with 768.0 GB of DDR5 memory, delivering a balanced 4.0 GB per core, and provides 50 Gbps of baseline network bandwidth. The server lacks local storage and GPU accelerators. In benchmarks, the m8g.metal-48xl delivers top-tier multi-core CPU performance and exceptional memory bandwidth, making it highly efficient for parallelized workloads. It excels in database operations, software compilation, and static web serving, as evidenced by its high Redis SET throughput and PassMark database scores. However, single-core performance is average, and the lack of local NVMe storage or GPUs limits its suitability for direct high-speed storage caching or heavy AI model training.

      Economics

      Average Price per Region

      Prices per Zone

      Lowest Prices

      Performance

      Workload Profiles

      9.4Score
      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) 35,079,899 ops/sec 1,918,343 ops/sec 18 50.00% 50.00% +320.00%
      Redis RPS (pipeline=16, SET) 217,912,339 ops/sec 11,704,734 ops/sec 18 20.00% 20.00% +78.00%
      PassMark Memory Mark (composite) 3,348 2,416 1.4 10.00% 10.00% +3.40%
      Memory bandwidth (read, 16 MB ~ L3) 896,810 MB/sec 108,420 MB/sec 8.2 10.00% 10.00% +23.00%
      PassMark single-thread CPU 1,873 Mops/s 2,365 Mops/s 0.79 10.00% 10.00% -2.30%

      Memory Bandwidth

      Compression

      OpenSSL

      Geekbench Single-Core

      Score: 1,667

      Geekbench Multi-Core

      Score: 20,860

      Passmark CPU Scores

      BENCHMARKSCORE
      Mark
      		 75,393
      Compression
      		 1,214,249
      Encryption
      		 31,878
      Extended Instructions
      		 138,526
      Floating Point Maths
      		 1,122,106
      Integer Maths
      		 713,112
      Physics
      		 38,059
      Prime Numbers
      		 1,682
      Single Threaded
      		 1,873
      String Sorting
      		 803,397

      Passmark Memory Scores

      BENCHMARKSCORE
      Memory Mark
      		 3,348
      Database Operations
      		 46,020
      Memory Latency
      		 46
      Memory Read Cached
      		 20,296
      Memory Read Uncached
      		 18,728
      Memory Write
      		 20,568

      Stress-ng Stressors

      CPU integerSingleMultiScaling
      atomic
      		 69.44 - -
      bitops
      		 139,686 26,483,347 189.59×
      branch
      		 219,318 33,532,589 152.89×
      funccall
      		 1,873 256,798 137.13×
      funcret
      		 24,550 3,686,279 150.15×
      hash
      		 150,279 22,797,940 151.70×
      intmath
      		 31,660 4,106,124 129.69×
      nop
      		 439,889 61,334,727 139.43×
      prime
      		 483,631 62,928,045 130.12×
      rotate
      		 128,035 16,691,536 130.37×
      str
      		 669,861 75,376,077 112.52×
      wcs
      		 193,234 31,056,184 160.72×
      CryptographySingleMultiScaling
      crypt
      		 3,402 466,863 137.22×
      CompressionSingleMultiScaling
      jpeg
      		 715.89 94,119 131.47×
      zlib
      		 45.75 7,029 153.65×
      HypervisorSingleMultiScaling
      clock
      		 37,608 147,617 3.93×
      clone
      		 349.11 19,910 57.03×
      context
      		 3,294 423,271 128.51×
      cpu-sched
      		 73.02 11,871 162.57×
      eventfd
      		 133,241 5,887,990 44.19×
      fault
      		 23,131 212,083 9.17×
      fork
      		 4,918 38,425 7.81×
      getrandom
      		 60,107 8,935,572 148.66×
      hrtimers
      		 126,114 10,127,381 80.30×
      mprotect
      		 652,406 21,383,900 32.78×
      mtx
      		 8,855,536 4,771,872,113 538.86×
      munmap
      		 3,686 68,108 18.48×
      null
      		 570,294 67,242,575 117.91×
      pipe
      		 1,048,387 57,967,808 55.29×
      priv-instr
      		 879,587 9,294,807 10.57×
      resched
      		 3,545,170 25,725,161 7.26×
      switch
      		 395,169 27,436,097 69.43×
      timermix
      		 757,257 124,665,273 164.63×
      tlb-shootdown
      		 10,894 105,626 9.70×
      vdso
      		 43,233,385 5,105,271,049 118.09×
      vfork
      		 43,538 194,241 4.46×
      yield
      		 80,378,906 88,077,668 1.10×
      zero
      		 2,702,288 491,239,117 181.79×
      CPU float/SIMDSingleMultiScaling
      eigen
      		 1,864 205,173 110.05×
      expmath
      		 354.09 37,837 106.86×
      fma
      		 641,371 74,646,856 116.39×
      fp
      		 146.08 22,998 157.44×
      fractal
      		 7.6 1,011 133.06×
      hyperbolic
      		 342.06 50,481 147.58×
      logmath
      		 414.66 68,936 166.25×
      matrix
      		 6,367 923,882 145.10×
      matrix-3d
      		 1,113 42,839 38.47×
      monte-carlo
      		 226.56 28,672 126.56×
      mpfr
      		 1,386 141,790 102.30×
      powmath
      		 240.46 31,390 130.54×
      trig
      		 459.34 56,253 122.46×
      veccmp
      		 21,065 2,835,267 134.60×
      vecfp
      		 1,018 134,386 132.02×
      vecmath
      		 4,291 523,265 121.94×
      vecshuf
      		 853.12 101,786 119.31×
      vecwide
      		 21,402 2,399,521 112.12×
      AlgorithmsSingleMultiScaling
      bitonicsort
      		 6.8 687.16 101.05×
      bsearch
      		 330.71 47,675 144.16×
      bubblesort
      		 1.2 147.4 122.83×
      fibsearch
      		 352.37 38,689 109.80×
      heapsort
      		 12.6 1,958 155.39×
      hsearch
      		 8,227 1,148,473 139.59×
      insertionsort
      		 6.6 822.39 124.60×
      malloc
      		 3,925,840 253,388,767 64.54×
      mcontend
      		 114.64 398,396 3475.19×
      mergesort
      		 32.2 3,348 103.97×
      qsort
      		 21.4 2,489 116.29×
      skiplist
      		 6,459 848,882 131.43×
      tree
      		 1.8 91.88 51.04×
      MemorySingleMultiScaling
      cache
      		 405,835 3,125,509 7.70×
      cachehammer
      		 1,838,556 28,123,537 15.30×
      cacheline
      		 84.16 5,760 68.44×
      l1cache
      		 319.11 51,711 162.05×
      memcpy
      		 630.02 95,690 151.88×
      memrate
      		 5.33 135.5 25.42×
      memthrash
      		 480.5 240.08 0.50×
      ptr-chase
      		 30,065,417 921,362,034 30.65×
      spinmem
      		 4,243 115,670 27.26×
      stream
      		 47.46 2,136,727 45021.64×
      vm
      		 110,546 5,399,411 48.84×

      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

      PostgreSQL heavy read-only throughput

      Alternatives

      Servers of the Same Family

      INSTANCEvCPUsMEMORYAccelerators
      m8g.medium 14 GiB0
      m8g.large 28 GiB0
      m8g.xlarge 416 GiB0
      m8g.2xlarge 832 GiB0
      m8g.4xlarge 1664 GiB0
      m8g.8xlarge 32128 GiB0
      m8g.12xlarge 48192 GiB0

      Similar Servers

      INSTANCEVENDORvCPUsMEMORYAccelerators

      m8g.metal-48xl FAQs