c9g.16xlarge by Amazon Web Services

c9g.16xlarge is a Compute optimized [AWS Graviton processors] Gen9 16xlarge server offered by Amazon Web Services with 64 vCPUs, 128 GiB of memory and 0 GB of storage. The pricing starts at 0.5925 USD per hour.
64 vCPU
128 GiB Memory
Spare SCore
126,149
(All-cores)
2,007
(Single-core)

Specifications

Server Metadata

Vendor ID
aws
Name
c9g.16xlarge
Description
Compute optimized [AWS Graviton processors] Gen9 16xlarge
Family
Compute Optimized
Series
c9g
Hw Virt
Average Time To Start
5.67
Status
active
Observed At
2026-11-14T19:09:28.102548

Availability

Region / IDSpotOndemand
(MY) / ap-southeast-5
1.3235 USD/h
2.7162 USD/h
Ohio (US) / us-east-2
0.811967 USD/h
2.7725 USD/h
Northern Virgina (US) / us-east-1
0.822975 USD/h
2.7821 USD/h
Oregon (US) / us-west-2
1.07415 USD/h
2.7821 USD/h
Seoul (KR) / ap-northeast-2
0.913533 USD/h
3.1309 USD/h
Frankfurt (DE) / eu-central-1
1.120033 USD/h
3.1635 USD/h
Singapore (SG) / ap-southeast-1
1.2492 USD/h
3.1962 USD/h
Tokyo (JP) / ap-northeast-1
1.287733 USD/h
3.4899 USD/h

Processor

vCPUs
64
Hypervisor
nitro
CPU Allocation
Dedicated
CPU Cores
64
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
4 MiB
CPU L1I Cache
64 KiB
CPU L1I Cache Total
4 MiB
CPU L2 Cache
2 MiB
CPU L2 Cache Total
128 MiB
CPU L3 Cache
48 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
62.8
Scalability
98.12

System Resources and Accelerators

Memory
Memory Amount
128 GiB
Memory Amount Actual
126 GiB
Accelerators
Accelerator Count
0
Accelerator Memory Min
0 MiB
Accelerator Memory Total
0 MiB
Accelerators
    Storage
    Storage Size
    0 GB
    Storages
      Compatible Storage IDs
      standard, gp2, gp3, st1, sc1, io1, io2
      Network
      Network Speed Baseline
      34 Gbps
      Network Speed Max
      34 Gbps
      Network Storage Speed Baseline
      24 Gbps
      Network Storage Speed Max
      24 Gbps
      Inbound Traffic
      0 GB/month
      Outbound Traffic
      0 GB/month
      IPv4
      0

      CPU and System Topology

      Server Description

      High-density 64-core ARM architecture providing dedicated parallel processing power, 34 Gbps networking, and top-tier multi-core compilation performance for scalable workloads.

      Amazon Web Services c9g.16xlarge is a compute-optimized instance equipped with 64 dedicated ARMv9 Neoverse-V3 physical cores operating at 3.3 GHz and 128.0 GB of RAM. Built on the AWS Nitro hypervisor without hyperthreading, it features a 1:2 core-to-memory ratio, a 48 MB L3 cache, baseline network bandwidth of 34 Gbps, and no local NVMe storage. Benchmarks place the system in the top tier for multi-core processing, software compilation, data decompression, Redis SET operations, and static web serving throughput. Read and write memory bandwidth performance rank in the top quartile, whereas single-core stress performance and small-to-medium model LLM inference speeds sit in the middle tier. By providing dense dedicated core allocation without hyperthreading contention or local disk overhead, the server offers efficient core scaling. Amazon Web Services c9g.16xlarge is effective for batch processing, CI/CD build clusters, web server fleets, and high-throughput key-value databases.

      Economics

      Average Price per Region

      Prices per Zone

      Lowest Prices

      Performance

      Workload Profiles

      6.9Score
      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) 21,899,341 ops/sec 1,933,252 ops/sec 12 50.00% 50.00% +250.00%
      Redis RPS (pipeline=16, SET) 128,991,809 ops/sec 11,814,424 ops/sec 11 20.00% 20.00% +62.00%
      PassMark Memory Mark (composite) 4,705 2,408 2 10.00% 10.00% +7.20%
      Memory bandwidth (read, 16 MB ~ L3) 440,996 MB/sec 108,674 MB/sec 4 10.00% 10.00% +15.00%
      PassMark single-thread CPU 2,463 Mops/s 2,365 Mops/s 1 10.00% 10.00% 0.00%

      Memory Bandwidth

      Compression

      OpenSSL

      Geekbench Single-Core

      Score: 2,328

      Geekbench Multi-Core

      Score: 29,666

      Passmark CPU Scores

      BENCHMARKSCORE
      Mark
      		 60,579
      Compression
      		 592,516
      Encryption
      		 18,932
      Extended Instructions
      		 85,477
      Floating Point Maths
      		 485,359
      Integer Maths
      		 388,330
      Physics
      		 18,879
      Prime Numbers
      		 1,087
      Single Threaded
      		 2,463
      String Sorting
      		 338,138

      Passmark Memory Scores

      BENCHMARKSCORE
      Memory Mark
      		 4,705
      Database Operations
      		 58,282
      Memory Latency
      		 26
      Memory Read Cached
      		 24,828
      Memory Read Uncached
      		 24,581
      Memory Write
      		 24,999

      Stress-ng Stressors

      CPU integerSingleMultiScaling
      atomic
      		 173.62 12.09 0.07×
      bitops
      		 165,220 10,434,528 63.16×
      branch
      		 320,251 20,087,900 62.73×
      funccall
      		 2,280 144,199 63.24×
      funcret
      		 40,259 2,522,542 62.66×
      hash
      		 180,906 11,486,045 63.49×
      intmath
      		 39,536 2,510,801 63.51×
      nop
      		 430,038 27,207,423 63.27×
      prime
      		 600,726 37,618,574 62.62×
      rotate
      		 149,189 9,420,638 63.15×
      str
      		 934,938 58,985,584 63.09×
      wcs
      		 272,641 17,573,687 64.46×
      CryptographySingleMultiScaling
      crypt
      		 4,361 270,088 61.94×
      CompressionSingleMultiScaling
      jpeg
      		 924.79 57,974 62.69×
      zlib
      		 45.77 2,939 64.22×
      HypervisorSingleMultiScaling
      clock
      		 49,064 333,285 6.79×
      clone
      		 413.32 14,163 34.27×
      context
      		 4,261 255,230 59.90×
      cpu-sched
      		 58.52 3,980 68.01×
      eventfd
      		 149,665 4,356,705 29.11×
      fault
      		 28,045 403,843 14.40×
      fork
      		 6,861 68,156 9.93×
      getrandom
      		 73,879 4,683,320 63.39×
      hrtimers
      		 1,400,566 5,066,567 3.62×
      mprotect
      		 1,190,045 20,025,285 16.83×
      mtx
      		 10,803,866 3,414,907,518 316.08×
      munmap
      		 4,862 109,856 22.59×
      null
      		 684,969 43,824,334 63.98×
      pipe
      		 1,368,455 38,375,870 28.04×
      priv-instr
      		 1,128,804 10,498,001 9.30×
      resched
      		 2,044,018 13,792,217 6.75×
      switch
      		 248,166 44,544,387 179.49×
      timermix
      		 591,899 10,817,938 18.28×
      tlb-shootdown
      		 17,213 269,129 15.64×
      vdso
      		 63,235,800 4,040,652,129 63.90×
      vfork
      		 56,514 384,060 6.80×
      yield
      		 134,553,909 123,001,685 0.91×
      zero
      		 3,261,439 208,706,586 63.99×
      CPU float/SIMDSingleMultiScaling
      eigen
      		 2,244 137,138 61.11×
      expmath
      		 425.27 26,599 62.55×
      fma
      		 852,495 53,970,927 63.31×
      fp
      		 177.04 11,176 63.13×
      fractal
      		 9.2 570.34 61.99×
      hyperbolic
      		 408.35 25,577 62.64×
      logmath
      		 491.27 31,021 63.14×
      matrix
      		 7,993 504,179 63.08×
      matrix-3d
      		 1,501 18,622 12.40×
      monte-carlo
      		 269.65 17,312 64.20×
      mpfr
      		 1,889 119,456 63.23×
      powmath
      		 282.08 17,855 63.30×
      trig
      		 553.81 35,071 63.33×
      veccmp
      		 26,347 1,668,210 63.32×
      vecfp
      		 1,860 117,719 63.30×
      vecmath
      		 4,810 305,075 63.43×
      vecshuf
      		 1,029 65,180 63.32×
      vecwide
      		 29,960 2,048,808 68.38×
      AlgorithmsSingleMultiScaling
      bitonicsort
      		 11.2 715.58 63.89×
      bsearch
      		 422.49 26,774 63.37×
      bubblesort
      		 1.4 89.6 64.00×
      fibsearch
      		 414.17 26,032 62.85×
      heapsort
      		 15.4 972.58 63.15×
      hsearch
      		 10,680 681,564 63.82×
      insertionsort
      		 8 494 61.75×
      malloc
      		 4,729,135 129,148,673 27.31×
      mcontend
      		 207.51 158,916 765.82×
      mergesort
      		 38.8 2,452 63.19×
      qsort
      		 26.6 1,692 63.60×
      skiplist
      		 7,669 482,582 62.93×
      tree
      		 2.2 76.93 34.97×
      MemorySingleMultiScaling
      cache
      		 573,911 1,981,236 3.45×
      cachehammer
      		 2,305,648 42,278,838 18.34×
      cacheline
      		 74.37 4,845 65.15×
      l1cache
      		 389.48 24,202 62.14×
      memcpy
      		 1,013 60,685 59.88×
      memrate
      		 6.33 61.99 9.79×
      memthrash
      		 274.61 458.47 1.67×
      ptr-chase
      		 19,430,599 699,531,307 36.00×
      spinmem
      		 5,541 71,214 12.85×
      stream
      		 53.49 58,553 1094.65×
      vm
      		 203,361 6,287,972 30.92×

      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
      c6g.medium 12 GiB0
      c6gd.medium 12 GiB0
      c6gn.medium 12 GiB0
      c7a.medium 12 GiB0
      c7g.medium 12 GiB0
      c7gd.medium 12 GiB0
      c7gn.medium 12 GiB0

      Similar Servers

      INSTANCEVENDORvCPUsMEMORYAccelerators

      c9g.16xlarge FAQs