n4-highcpu-64 by Google Cloud Platform

n4-highcpu-64 is a 64 vCPUs, 128 GB RAM server offered by Google Cloud Platform with 64 vCPUs, 128 GiB of memory and 0 GB of storage. The pricing starts at 0.3092 USD per hour.
64 vCPU
128 GiB Memory
Spare SCore
58,262
(All-cores)
1,786
(Single-core)

Specifications

Server Metadata

Vendor ID
gcp
Server ID
1750023
Name
n4-highcpu-64
Description
64 vCPUs, 128 GB RAM
Family
n4
Hw Virt
Average Time To Start
20
Status
active
Observed At
2026-11-02T12:29:21.240508

Availability

REGION / IDSPOTONDEMAND
Columbus (US) / us-east5
0.8215 USD/h 2.4493 USD/h
Ashburn (US) / us-east4
1.2438 USD/h 2.4493 USD/h
Council Bluffs (US) / us-central1
1.3765 USD/h 2.4493 USD/h
The Dalles (US) / us-west1
1.3765 USD/h 2.4493 USD/h
Moncks Corner (US) / us-east1
1.3765 USD/h 2.4493 USD/h
Mumbai (IN) / asia-south1
0.5536 USD/h 2.5473 USD/h
Delhi (IN) / asia-south2
0.8154 USD/h 2.5473 USD/h
Eemshaven (NL) / europe-west4
0.5371 USD/h 2.5717 USD/h
Stockholm (SE) / europe-north2
1.0601 USD/h 2.5717 USD/h
Bangkok (TH) / asia-southeast3
0.4869 USD/h 2.5718 USD/h
Queretaro (MX) / northamerica-south1
0.7163 USD/h 2.6697 USD/h
Tel Aviv (IL) / me-west1
0.7064 USD/h 2.6942 USD/h
St. Ghislain (BE) / europe-west1
0.352 USD/h 2.6969 USD/h
Montréal (CA) / northamerica-northeast1
0.5032 USD/h 2.6969 USD/h
Johannesburg (ZA) / africa-south1
0.647 USD/h 2.6969 USD/h
Toronto (CA) / northamerica-northeast2
0.993 USD/h 2.6969 USD/h
Hamina (FI) / europe-north1
1.1186 USD/h 2.6969 USD/h
Las Vegas (US) / us-west4
1.1514 USD/h 2.7586 USD/h
London (GB) / europe-west2
1.0698 USD/h 2.7922 USD/h
Changhua County (TW) / asia-east1
1.0781 USD/h 2.836 USD/h
Milan (IT) / europe-west8
0.5257 USD/h 2.8412 USD/h
Paris (FR) / europe-west9
1.0162 USD/h 2.8412 USD/h
Dallas (US) / us-south1
0.508 USD/h 2.8902 USD/h
Madrid (ES) / europe-southwest1
0.6445 USD/h 2.8902 USD/h
Frankfurt (DE) / europe-west3
1.7213 USD/h 2.8902 USD/h
Salt Lake City (US) / us-west3
0.6263 USD/h 2.9443 USD/h
Los Angeles (US) / us-west2
1.668 USD/h 2.9443 USD/h
Warsaw (PL) / europe-central2
1.506 USD/h 2.9636 USD/h
Doha (QA) / me-central1
1.3688 USD/h 2.9759 USD/h
Jurong West (SG) / asia-southeast1
1.8118 USD/h 3.0217 USD/h
Sydney (AU) / australia-southeast1
1.4826 USD/h 3.0616 USD/h
Osaka (JP) / asia-northeast2
1.1395 USD/h 3.1454 USD/h
Seoul (KR) / asia-northeast3
1.4854 USD/h 3.1454 USD/h
Tokyo (JP) / asia-northeast1
1.7882 USD/h 3.1454 USD/h
Melbourne (AU) / australia-southeast2
1.1215 USD/h 3.1841 USD/h
Zurich (CH) / europe-west6
1.6307 USD/h 3.2086 USD/h
Jakarta (ID) / asia-southeast2
0.9549 USD/h 3.2935 USD/h
Hong Kong / asia-east2
0.3092 USD/h 3.4263 USD/h
Santiago (CL) / southamerica-west1
0.6487 USD/h 3.5025 USD/h
Osasco (BR) / southamerica-east1
0.5135 USD/h 3.8892 USD/h
Dammam (SA) / me-central2
0.7105 USD/h 3.9188 USD/h

Processor

vCPUs
64
CPU Allocation
Dedicated
CPU Cores
32
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU L1D Cache
48 KiB
CPU L1D Cache Total
2 MiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
1 MiB
CPU L2 Cache
2 MiB
CPU L2 Cache Total
64 MiB
CPU L3 Cache
260 MiB
CPU L3 Cache Total
260 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, ss, ht, syscall, nx, pdpe1gb, rdtscp, lm, constant_tsc, rep_good, nopl, xtopology, nonstop_tsc, cpuid, tsc_known_freq, pni, pclmulqdq, smx, est, ssse3, sdbg, fma, cx16, xtpr, pdcm, pcid, dca, sse4_1, sse4_2, x2apic, movbe, popcnt, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, abm, 3dnowprefetch, cat_l3, cat_l2, cdp_l3, cdp_l2, ssbd, mba, ibrs, ibpb, stibp, ibrs_enhanced, fsgsbase, tsc_adjust, bmi1, hle, avx2, smep, bmi2, erms, invpcid, rtm, avx512f, avx512dq, rdseed, adx, smap, avx512ifma, clflushopt, clwb, avx512cd, sha_ni, avx512bw, avx512vl, xsaveopt, xsavec, xgetbv1, xsaves, cqm_llc, cqm_occup_llc, cqm_mbm_total, cqm_mbm_local, avx_vnni, avx512_bf16, wbnoinvd, arat, avx512vbmi, umip, avx512_vbmi2, gfni, vaes, vpclmulqdq, avx512_vnni, avx512_bitalg, avx512_vpopcntdq, la57, rdpid, cldemote, movdiri, movdir64b, fsrm, md_clear, serialize, tsxldtrk, amx_bf16, avx512_fp16, amx_tile, amx_int8, arch_capabilities
Ecpus
32.6
Scalability
50.94

System Resources and Accelerators

MEMORY
Memory Amount
128 GiB
Memory Amount Actual
128 GiB
ACCELERATORS
Accelerator Count
0
Accelerator Memory Min
0 MiB
Accelerator Memory Total
0 MiB
Accelerators
    STORAGE
    Storage Size
    0 GB
    Storages
      NETWORK
      Inbound Traffic
      0 GB/month
      Outbound Traffic
      0 GB/month
      IPv4
      0

      CPU and System Topology

      Server Description

      A high-density compute server featuring sixty-four dedicated cores and advanced matrix extensions for accelerated processing and database workloads.

      Compute OptimizedGeneral Purpose

      Google Cloud Platform n4-highcpu-64 is a compute-optimized x86_64 server featuring 64 dedicated vCPUs across 32 physical Intel cores and 128 GB of RAM. Lacking local storage and GPU accelerators, this instance relies on its high-density CPU configuration and a 260 MB L3 cache. It supports advanced instruction sets like AVX-512 and Intel AMX. Benchmark results place the server in the top tier for database operations and image processing, with strong performance in software compilation, ray tracing, and multi-core processing. It also exhibits strong throughput for Redis SET operations and static web serving. The 2.0 GB of RAM per core makes it highly efficient for compute-bound tasks, though it is less suited for memory-heavy applications. This server is recommended for high-throughput web serving, software compilation, cryptography, and CPU-based machine learning inference.

      Economics

      Average Price per Region

      Prices per Zone

      Lowest Prices

      Performance

      Workload Profiles

      3.3Score
      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) 9,557,191 ops/sec 1,918,343 ops/sec 5.1 50.00% 50.00% +130.00%
      Redis RPS (pipeline=16, SET) 49,540,796 ops/sec 11,704,734 ops/sec 4.2 20.00% 20.00% +33.00%
      PassMark Memory Mark (composite) 2,616 2,416 1.1 10.00% 10.00% +0.96%
      Memory bandwidth (read, 16 MB ~ L3) 241,127 MB/sec 108,466 MB/sec 2.2 10.00% 10.00% +8.20%
      PassMark single-thread CPU 2,711 Mops/s 2,365 Mops/s 1.1 10.00% 10.00% +0.96%

      Memory Bandwidth

      Compression

      OpenSSL

      Geekbench Single-Core

      Score: 1,881

      Geekbench Multi-Core

      Score: 18,694

      Passmark CPU Scores

      BENCHMARKSCORE
      Mark
      		 69,148
      Compression
      		 1,008,183
      Encryption
      		 60,374
      Extended Instructions
      		 75,345
      Floating Point Maths
      		 191,088
      Integer Maths
      		 259,251
      Physics
      		 11,651
      Prime Numbers
      		 561
      Single Threaded
      		 2,711
      String Sorting
      		 135,188

      Passmark Memory Scores

      BENCHMARKSCORE
      Memory Mark
      		 2,616
      Database Operations
      		 33,232
      Memory Latency
      		 58
      Memory Read Cached
      		 23,822
      Memory Read Uncached
      		 10,958
      Memory Write
      		 10,188

      Stress-ng Stressors

      CPU integerSingleMultiScaling
      atomic
      		 206.39 128.24 0.62×
      bitops
      		 130,457 4,665,507 35.76×
      branch
      		 123,167 5,608,042 45.53×
      funccall
      		 659.46 25,191 38.20×
      funcret
      		 10,897 509,358 46.74×
      hash
      		 147,403 6,708,381 45.51×
      intmath
      		 56,403 2,073,895 36.77×
      nop
      		 235,727 5,628,793 23.88×
      prime
      		 422,249 15,054,845 35.65×
      rotate
      		 88,285 2,744,056 31.08×
      str
      		 798,497 35,116,559 43.98×
      wcs
      		 601,188 22,809,851 37.94×
      CryptographySingleMultiScaling
      crypt
      		 2,896 101,887 35.19×
      CompressionSingleMultiScaling
      jpeg
      		 699.04 27,847 39.84×
      zlib
      		 54.36 2,468 45.41×
      HypervisorSingleMultiScaling
      clock
      		 34,715 429,424 12.37×
      clone
      		 436.47 12,327 28.24×
      context
      		 4,535 223,510 49.28×
      cpu-sched
      		 60.63 4,144 68.35×
      eventfd
      		 58,162 1,845,327 31.73×
      fault
      		 23,085 271,948 11.78×
      fork
      		 4,518 107,011 23.68×
      getrandom
      		 50,524 2,244,481 44.42×
      hrtimers
      		 614,100 3,765,110 6.13×
      mprotect
      		 1,032,981 15,555,545 15.06×
      mtx
      		 13,130,618 1,803,518,987 137.35×
      munmap
      		 3,306 102,713 31.07×
      null
      		 798,258 36,668,579 45.94×
      pipe
      		 649,128 49,845,644 76.79×
      priv-instr
      		 733,310 14,971,856 20.42×
      resched
      		 14,528,928 7,290,977 0.50×
      switch
      		 87,601 16,784,619 191.60×
      timermix
      		 207,443 10,526,718 50.75×
      tlb-shootdown
      		 12,448 120,690 9.70×
      tsc
      		 518,793 25,191,412 48.56×
      usersyscall
      		 624,643 14,852,765 23.78×
      vdso
      		 79,841,830 3,621,920,265 45.36×
      vfork
      		 28,977 232,681 8.03×
      x86syscall
      		 8,846,668 436,278,128 49.32×
      yield
      		 51,633,516 65,039,659 1.26×
      zero
      		 5,759,073 268,180,384 46.57×
      CPU float/SIMDSingleMultiScaling
      eigen
      		 13,958 549,559 39.37×
      expmath
      		 3,061 132,864 43.40×
      fma
      		 1,124,587 44,176,360 39.28×
      fp
      		 160.14 7,835 48.93×
      fractal
      		 5.6 242.71 43.34×
      hyperbolic
      		 3,503 151,842 43.34×
      logmath
      		 4,484 213,298 47.57×
      matrix
      		 4,294 150,158 34.97×
      matrix-3d
      		 780.67 9,988 12.79×
      monte-carlo
      		 248.39 12,406 49.95×
      mpfr
      		 1,553 52,961 34.10×
      powmath
      		 1,746 77,527 44.40×
      trig
      		 2,370 91,473 38.60×
      veccmp
      		 21,913 752,431 34.34×
      vecfp
      		 1,343 50,131 37.33×
      vecmath
      		 3,448 142,694 41.38×
      vecshuf
      		 729.62 24,520 33.61×
      vecwide
      		 29,895 1,090,269 36.47×
      AlgorithmsSingleMultiScaling
      bitonicsort
      		 7.8 269.79 34.59×
      bsearch
      		 250.72 10,423 41.57×
      bubblesort
      		 0.8 25.6 32.00×
      fibsearch
      		 271.76 9,493 34.93×
      heapsort
      		 8.6 347.19 40.37×
      hsearch
      		 7,354 269,678 36.67×
      insertionsort
      		 11 288.8 26.25×
      malloc
      		 2,269,671 77,142,089 33.99×
      mcontend
      		 93.92 10,286 109.52×
      mergesort
      		 27.2 1,140 41.91×
      qsort
      		 19.4 802.78 41.38×
      skiplist
      		 4,750 223,485 47.04×
      tree
      		 1.4 51.11 36.51×
      MemorySingleMultiScaling
      cache
      		 210,912 3,563,360 16.89×
      cachehammer
      		 195,842 6,134,352 31.32×
      cacheline
      		 51.57 3,335 64.67×
      l1cache
      		 946.99 10,617 11.21×
      memcpy
      		 391.52 11,408 29.14×
      memrate
      		 0.18 9.31 51.72×
      memthrash
      		 273.69 1,746 6.38×
      ptr-chase
      		 25,579,499 571,637,824 22.35×
      spinmem
      		 223.59 2,782 12.44×
      stream
      		 5.4 6,403 1185.78×
      vm
      		 120,456 301,445 2.50×

      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
      n4-highcpu-2 24 GiB0
      n4-standard-2 28 GiB0
      n4-highmem-2 216 GiB0
      n4-highcpu-4 48 GiB0
      n4-standard-4 416 GiB0
      n4-highmem-4 432 GiB0
      n4-highcpu-8 816 GiB0

      Similar Servers

      INSTANCEVENDORvCPUsMEMORYAccelerators

      n4-highcpu-64 FAQs