c3d-highmem-90 by Google Cloud Platform

c3d-highmem-90 is a 90 vCPUs, 720 GB RAM server offered by Google Cloud Platform with 90 vCPUs, 720 GiB of memory and 0 GB of storage. The pricing starts at 1.2613 USD per hour.
90 vCPU
720 GiB Memory
Spare SCore
129,141
(All-cores)
2,866
(Single-core)

Specifications #

Server Metadata #

Vendor ID
gcp
Server ID
1710090
Name
c3d-highmem-90
Description
90 vCPUs, 720 GB RAM
Family
General Purpose
Series
c3d
Hw Virt
Average Time To Start
64.67
Status
active
Observed At
2026-11-20T02:13:47.371293

Availability #

Region / IDSpotOndemand
Council Bluffs (US) / us-central1
1.2613 USD/h
5.5111 USD/h
The Dalles (US) / us-west1
1.2613 USD/h
5.5111 USD/h
Moncks Corner (US) / us-east1
1.2613 USD/h
5.5111 USD/h
Ashburn (US) / us-east4
2.7268 USD/h
5.5111 USD/h
Mumbai (IN) / asia-south1
1.3437 USD/h
5.7316 USD/h
Stockholm (SE) / europe-north2
2.4514 USD/h
5.7867 USD/h
Eemshaven (NL) / europe-west4
3.4715 USD/h
5.7867 USD/h
Queretaro (MX) / northamerica-south1
1.3277 USD/h
6.0072 USD/h
Tel Aviv (IL) / me-west1
3.6346 USD/h
6.0623 USD/h
St. Ghislain (BE) / europe-west1
1.8745 USD/h
6.0683 USD/h
Hamina (FI) / europe-north1
3.0665 USD/h
6.0683 USD/h
Columbus (US) / us-east5
1.8736 USD/h
6.2072 USD/h
London (GB) / europe-west2
2.6068 USD/h
6.2827 USD/h
Changhua County (TW) / asia-east1
2.6302 USD/h
6.3814 USD/h
Frankfurt (DE) / europe-west3
3.697 USD/h
6.5032 USD/h
Jurong West (SG) / asia-southeast1
3.8588 USD/h
6.7991 USD/h
Dammam (SA) / me-central2
1.4861 USD/h
8.8178 USD/h

Processor #

vCPUs
90
CPU Allocation
Dedicated
CPU Cores
45
CPU Architecture
x86_64
CPU Manufacturer
AMD
CPU Family
EPYC
CPU Model
9B14
CPU L1D Cache
32 KiB
CPU L1D Cache Total
1 MiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
1 MiB
CPU L2 Cache
1 MiB
CPU L2 Cache Total
45 MiB
CPU L3 Cache
32 MiB
CPU L3 Cache Total
192 MiB
CPU Flags
fpu, vme, de, pse, tsc, msr, pae, mce, cx8, apic, sep, mtrr, pge, mca, cmov, pat, pse36, clflush, mmx, fxsr, sse, sse2, ht, syscall, nx, mmxext, fxsr_opt, pdpe1gb, rdtscp, lm, constant_tsc, rep_good, nopl, nonstop_tsc, cpuid, extd_apicid, tsc_known_freq, pni, pclmulqdq, ssse3, fma, cx16, pcid, sse4_1, sse4_2, x2apic, movbe, popcnt, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, cmp_legacy, cr8_legacy, abm, sse4a, misalignsse, 3dnowprefetch, osvw, topoext, ssbd, ibrs, ibpb, stibp, vmmcall, fsgsbase, tsc_adjust, bmi1, avx2, smep, bmi2, erms, invpcid, avx512f, avx512dq, rdseed, adx, smap, avx512ifma, clflushopt, clwb, avx512cd, sha_ni, avx512bw, avx512vl, xsaveopt, xsavec, xgetbv1, xsaves, avx512_bf16, clzero, xsaveerptr, wbnoinvd, arat, avx512vbmi, umip, avx512_vbmi2, gfni, vaes, vpclmulqdq, avx512_vnni, avx512_bitalg, avx512_vpopcntdq, rdpid, fsrm
Ecpus
45.1
Scalability
50.11

System Resources and Accelerators #

Memory
Memory Amount
720 GiB
Memory Amount Actual
720 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
      30014, 30017, 30012, 30013, 30016, 30007, 30002
      Network
      Inbound Traffic
      0 GB/month
      Outbound Traffic
      0 GB/month
      IPv4
      0

      CPU and System Topology #

      Server Description #

      A memory-optimized x86_64 platform featuring ninety dedicated vCPUs and seven hundred twenty gigabytes of RAM for memory-intensive workloads.

      Google Cloud Platform c3d-highmem-90 is a memory-optimized server featuring 90 dedicated vCPUs and 720.0 GB of RAM, powered by an AMD EPYC 9B14 processor on the x86_64 architecture. It provides a high resource density of 8.0 GB of memory per core. While the instance lacks local storage and GPU accelerators, it delivers top-tier multi-core CPU performance, ranking in the top 10% on stress-ng and PassMark benchmarks. Memory read bandwidth is strong at 218.2 GB/sec for large blocks, though write bandwidth is average at 77.77 GB/sec. Single-core performance also falls in the average tier. This hardware profile is suited for memory-intensive workloads, large-scale database operations, and multi-threaded software compilation, offering a balanced tradeoff between massive memory capacity and high multi-core compute density.

      Economics #

      Average Price per Region #

      Prices per Zone #

      Lowest Prices #

      Performance #

      Workload Profiles #

      3.4Score
      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) 8,678,839 ops/sec 1,936,298 ops/sec 4.6 50.00% 50.00% +110.00%
      Redis RPS (pipeline=16, SET) 65,834,413 ops/sec 11,817,332 ops/sec 5.5 20.00% 20.00% +41.00%
      PassMark Memory Mark (composite) 2,894 2,408 1.2 10.00% 10.00% +1.80%
      Memory bandwidth (read, 16 MB ~ L3) 244,867 MB/sec 108,674 MB/sec 2.2 10.00% 10.00% +8.20%
      PassMark single-thread CPU 2,698 Mops/s 2,365 Mops/s 1.1 10.00% 10.00% +0.96%

      Memory Bandwidth#

      Compression #

      OpenSSL #

      Geekbench Single-Core #

      Score: 1,877

      Geekbench Multi-Core #

      Score: 19,539

      Passmark CPU Scores #

      BENCHMARKSCORE
      Mark
      87,517
      Compression
      1,620,291
      Encryption
      100,770
      Extended Instructions
      115,339
      Floating Point Maths
      228,290
      Integer Maths
      398,253
      Physics
      12,390
      Prime Numbers
      693
      Single Threaded
      2,698
      String Sorting
      224,047

      Passmark Memory Scores #

      BENCHMARKSCORE
      Memory Mark
      2,894
      Database Operations
      28,335
      Memory Latency
      70
      Memory Read Cached
      25,826
      Memory Read Uncached
      24,970
      Memory Write
      18,526

      Stress-ng Stressors #

      CPU integerSingleMultiScaling
      atomic
      771.67 180.76 0.23×
      bitops
      148,383 7,994,564 53.88×
      branch
      197,523 15,440,082 78.17×
      funccall
      574.02 35,441 61.74×
      funcret
      19,513 1,276,126 65.40×
      hash
      147,401 9,959,787 67.57×
      intmath
      61,361 3,747,017 61.06×
      nop
      584,499 25,847,789 44.22×
      prime
      451,618 26,564,160 58.82×
      rotate
      82,006 3,669,480 44.75×
      str
      984,018 54,408,630 55.29×
      wcs
      637,876 33,295,065 52.20×
      CryptographySingleMultiScaling
      crypt
      3,491 169,182 48.46×
      CompressionSingleMultiScaling
      jpeg
      786.74 43,110 54.80×
      zlib
      45.44 4,499 99.00×
      HypervisorSingleMultiScaling
      clock
      14,719 237,242 16.12×
      clone
      222.95 7,646 34.29×
      context
      1,728 106,845 61.82×
      cpu-sched
      66.71 5,764 86.40×
      eventfd
      35,593 3,622,522 101.78×
      fault
      10,301 217,843 21.15×
      fork
      3,341 82,386 24.66×
      getrandom
      47,538 2,779,287 58.47×
      hrtimers
      247,641 1,942,967 7.85×
      mprotect
      647,373 11,914,923 18.41×
      mtx
      41,676,547 3,865,048,511 92.74×
      munmap
      2,477 83,426 33.68×
      null
      182,192 11,385,946 62.49×
      pipe
      931,735 23,739,664 25.48×
      priv-instr
      318,285 18,025,835 56.63×
      resched
      4,968,449 3,656,118 0.74×
      switch
      121,593 12,024,030 98.89×
      timermix
      151,138 1,991,816 13.18×
      tlb-shootdown
      12,646 190,338 15.05×
      tsc
      731,021 60,683,216 83.01×
      usersyscall
      224,280 13,928,232 62.10×
      vdso
      89,221,997 6,531,000,156 73.20×
      vfork
      22,362 172,594 7.72×
      x86syscall
      2,128,158 128,922,385 60.58×
      yield
      39,896,046 40,690,907 1.02×
      zero
      1,622,328 104,163,556 64.21×
      CPU float/SIMDSingleMultiScaling
      eigen
      15,253 731,511 47.96×
      expmath
      2,787 198,169 71.10×
      fma
      853,754 63,031,615 73.83×
      fp
      234.13 16,100 68.77×
      fractal
      7 528.82 75.55×
      hyperbolic
      3,312 225,866 68.20×
      logmath
      4,238 273,918 64.63×
      matrix
      4,588 242,880 52.94×
      matrix-3d
      893.14 7,162 8.02×
      monte-carlo
      106.76 6,499 60.87×
      mpfr
      1,564 90,042 57.59×
      powmath
      1,251 91,120 72.85×
      trig
      1,740 119,382 68.60×
      veccmp
      17,678 1,007,808 57.01×
      vecfp
      1,682 110,860 65.91×
      vecmath
      3,732 271,412 72.73×
      vecshuf
      593.58 32,727 55.13×
      vecwide
      26,576 1,557,669 58.61×
      AlgorithmsSingleMultiScaling
      bitonicsort
      9.6 490.58 51.10×
      bsearch
      283.69 17,590 62.01×
      bubblesort
      0.8 54 67.50×
      fibsearch
      353.45 18,990 53.73×
      heapsort
      10.2 619.59 60.74×
      hsearch
      9,703 475,325 48.99×
      insertionsort
      11.8 547.79 46.42×
      malloc
      3,518,711 141,654,730 40.26×
      mcontend
      105.5 14,940 141.61×
      mergesort
      31.4 2,089 66.51×
      qsort
      20.2 1,283 63.49×
      skiplist
      6,536 445,318 68.13×
      tree
      1.8 67.46 37.48×
      MemorySingleMultiScaling
      cache
      798,927 1,348,461 1.69×
      cachehammer
      280,599 14,876,194 53.02×
      cacheline
      52.98 4,118 77.73×
      l1cache
      238.77 12,059 50.50×
      memcpy
      416.35 19,237 46.20×
      memrate
      2.72 15.64 5.75×
      memthrash
      302.45 1,777 5.88×
      ptr-chase
      43,877,948 685,697,993 15.63×
      spinmem
      257.79 5,165 20.04×
      stream
      114.21 88,999 779.26×
      vm
      58,570 303,889 5.19×

      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
      c3d-highcpu-4 48 GiB0
      c3d-standard-4 416 GiB0
      c3d-highmem-4 432 GiB0
      c3d-highcpu-8 816 GiB0
      c3d-standard-8 832 GiB0
      c3d-standard-8-lssd 832 GiB0
      c3d-highmem-8 864 GiB0

      Similar Servers #

      INSTANCEVENDORvCPUsMEMORYAccelerators

      c3d-highmem-90 FAQs #