n2-highcpu-96 by Google Cloud Platform

n2-highcpu-96 is a 96 vCPUs 96 GB RAM server offered by Google Cloud Platform with 96 vCPUs, 96 GiB of memory and 0 GB of storage. The pricing starts at 0.3408 USD per hour.
96 vCPU
96 GiB Memory
Spare SCore
90,557
(All-cores)
1,843
(Single-core)

Specifications

Server Metadata

Vendor ID
gcp
Server ID
903096
Name
n2-highcpu-96
Description
96 vCPUs 96 GB RAM
Family
n2
Hw Virt
Average Time To Start
18
Status
active
Observed At
2026-10-22T10:57:18.753991

Availability

REGION / IDSPOTONDEMAND
Council Bluffs (US) / us-central1
2.0642 USD/h 3.4414 USD/h
The Dalles (US) / us-west1
2.0642 USD/h 3.4414 USD/h
Moncks Corner (US) / us-east1
2.0642 USD/h 3.4414 USD/h
Columbus (US) / us-east5
2.0652 USD/h 3.4414 USD/h
Tel Aviv (IL) / me-west1
2.27 USD/h 3.7855 USD/h
St. Ghislain (BE) / europe-west1
0.8577 USD/h 3.7857 USD/h
Johannesburg (ZA) / africa-south1
1.0748 USD/h 3.7857 USD/h
Toronto (CA) / northamerica-northeast2
0.3408 USD/h 3.7887 USD/h
Eemshaven (NL) / europe-west4
1.0527 USD/h 3.7887 USD/h
Montréal (CA) / northamerica-northeast1
1.7364 USD/h 3.7887 USD/h
Hamina (FI) / europe-north1
1.8398 USD/h 3.7891 USD/h
Las Vegas (US) / us-west4
2.1501 USD/h 3.8759 USD/h
Ashburn (US) / us-east4
0.9676 USD/h 3.8761 USD/h
Changhua County (TW) / asia-east1
1.3796 USD/h 3.9848 USD/h
Milan (IT) / europe-west8
0.392 USD/h 3.992 USD/h
Paris (FR) / europe-west9
1.1162 USD/h 3.992 USD/h
Dallas (US) / us-south1
0.7632 USD/h 4.0609 USD/h
Madrid (ES) / europe-southwest1
1.0474 USD/h 4.0609 USD/h
Delhi (IN) / asia-south2
0.3723 USD/h 4.1336 USD/h
Mumbai (IN) / asia-south1
1.4348 USD/h 4.1336 USD/h
Los Angeles (US) / us-west2
1.3947 USD/h 4.1337 USD/h
Salt Lake City (US) / us-west3
1.5811 USD/h 4.1337 USD/h
Melbourne (AU) / australia-southeast2
1.9281 USD/h 4.1508 USD/h
Warsaw (PL) / europe-central2
1.3632 USD/h 4.1642 USD/h
Doha (QA) / me-central1
0.825 USD/h 4.1813 USD/h
Jurong West (SG) / asia-southeast1
1.6177 USD/h 4.2456 USD/h
Osaka (JP) / asia-northeast2
1.2312 USD/h 4.4195 USD/h
Seoul (KR) / asia-northeast3
1.2811 USD/h 4.4195 USD/h
Tokyo (JP) / asia-northeast1
1.2786 USD/h 4.4196 USD/h
London (GB) / europe-west2
0.8842 USD/h 4.434 USD/h
Frankfurt (DE) / europe-west3
2.4428 USD/h 4.434 USD/h
Berlin (DE) / europe-west10
0.6224 USD/h 4.4342 USD/h
Turin (IT) / europe-west12
0.8228 USD/h 4.4394 USD/h
Zurich (CH) / europe-west6
1.645 USD/h 4.5082 USD/h
Jakarta (ID) / asia-southeast2
2.6164 USD/h 4.6277 USD/h
Hong Kong / asia-east2
1.3576 USD/h 4.8153 USD/h
Sydney (AU) / australia-southeast1
2.2359 USD/h 4.8832 USD/h
Santiago (CL) / southamerica-west1
1.2615 USD/h 4.9212 USD/h
Osasco (BR) / southamerica-east1
1.3259 USD/h 5.4629 USD/h
Dammam (SA) / me-central2
0.9123 USD/h 5.5063 USD/h

Processor

vCPUs
96
CPU Allocation
Dedicated
CPU Cores
48
CPU Speed
2.6 GHz
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU Family
Xeon
CPU L1D Cache
48 KiB
CPU L1D Cache Total
2 MiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
2 MiB
CPU L2 Cache
1 MiB
CPU L2 Cache Total
60 MiB
CPU L3 Cache
54 MiB
CPU L3 Cache Total
108 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, ssse3, fma, cx16, pcid, sse4_1, sse4_2, x2apic, movbe, popcnt, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, abm, 3dnowprefetch, ssbd, 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, arat, avx512vbmi, umip, avx512_vbmi2, gfni, vaes, vpclmulqdq, avx512_vnni, avx512_bitalg, avx512_vpopcntdq, rdpid, fsrm, md_clear, arch_capabilities
Ecpus
49.1
Scalability
51.15

System Resources and Accelerators

MEMORY
Memory Amount
96 GiB
Memory Amount Actual
96 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-core-density compute server featuring ninety-six dedicated vCPUs and AVX-512 support for highly parallelized, CPU-bound workloads.

      Compute OptimizedGeneral Purpose

      Google Cloud Platform n2-highcpu-96 is an x86_64 compute-optimized server featuring 96 dedicated vCPUs across 48 physical Intel Xeon cores running at 2.6 GHz. It is configured with 96.0 GB of RAM, yielding a 1.0 GB per core memory ratio, with no bundled local storage or GPUs. The architecture supports AVX-512 instructions, enabling strong multi-core CPU performance, software compilation, and ray tracing capabilities that rank in the top 25% of benchmarks. While single-core performance remains average, the server excels in memory bandwidth and parallel processing. This instance is highly suited for compute-bound workloads, including software compilation, image processing, static web serving, and database operations, though its low memory-to-core ratio presents a tradeoff for memory-intensive applications.

      Economics

      Average Price per Region

      Prices per Zone

      Lowest Prices

      Performance

      Workload Profiles

      3.9Score
      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) 11,999,223 ops/sec 1,918,343 ops/sec 6.3 50.00% 50.00% +150.00%
      Redis RPS (pipeline=16, SET) 60,020,256 ops/sec 11,704,734 ops/sec 5 20.00% 20.00% +38.00%
      PassMark Memory Mark (composite) 2,655 2,416 1.1 10.00% 10.00% +0.96%
      Memory bandwidth (read, 16 MB ~ L3) 332,061 MB/sec 108,420 MB/sec 3 10.00% 10.00% +12.00%
      PassMark single-thread CPU 2,354 Mops/s 2,365 Mops/s 1 10.00% 10.00% 0.00%

      Memory Bandwidth

      Compression

      OpenSSL

      Geekbench Single-Core

      Score: 1,475

      Geekbench Multi-Core

      Score: 15,886

      Passmark CPU Scores

      BENCHMARKSCORE
      Mark
      		 73,075
      Compression
      		 1,393,128
      Encryption
      		 81,436
      Extended Instructions
      		 78,876
      Floating Point Maths
      		 199,905
      Integer Maths
      		 391,103
      Physics
      		 7,116
      Prime Numbers
      		 422
      Single Threaded
      		 2,354
      String Sorting
      		 182,794

      Passmark Memory Scores

      BENCHMARKSCORE
      Memory Mark
      		 2,655
      Database Operations
      		 29,771
      Memory Latency
      		 54
      Memory Read Cached
      		 23,561
      Memory Read Uncached
      		 10,731
      Memory Write
      		 10,323

      Stress-ng Stressors

      CPU integerSingleMultiScaling
      atomic
      		 224.26 248.7 1.11×
      bitops
      		 126,848 6,191,871 48.81×
      branch
      		 133,506 10,160,212 76.10×
      funccall
      		 602.08 28,894 47.99×
      funcret
      		 10,751 719,430 66.92×
      hash
      		 146,501 10,027,411 68.45×
      intmath
      		 59,467 3,479,002 58.50×
      nop
      		 183,667 7,813,590 42.54×
      prime
      		 383,375 18,901,191 49.30×
      rotate
      		 95,033 4,871,820 51.26×
      str
      		 872,481 38,829,974 44.51×
      wcs
      		 597,663 22,978,764 38.45×
      CryptographySingleMultiScaling
      crypt
      		 2,766 122,685 44.36×
      CompressionSingleMultiScaling
      jpeg
      		 686.85 38,358 55.85×
      zlib
      		 62.42 3,801 60.90×
      HypervisorSingleMultiScaling
      clock
      		 33,318 237,233 7.12×
      clone
      		 461.5 12,138 26.30×
      context
      		 4,563 333,929 73.19×
      cpu-sched
      		 57.77 5,639 97.61×
      eventfd
      		 58,033 2,662,920 45.89×
      fault
      		 15,840 220,323 13.91×
      fork
      		 4,688 54,670 11.66×
      getrandom
      		 48,152 2,555,581 53.07×
      hrtimers
      		 586,871 11,691,163 19.92×
      mprotect
      		 744,217 16,181,978 21.74×
      mtx
      		 13,092,277 2,554,799,142 195.14×
      munmap
      		 3,272 66,673 20.38×
      null
      		 720,282 42,270,585 58.69×
      pipe
      		 821,805 46,786,279 56.93×
      priv-instr
      		 737,664 14,992,875 20.32×
      resched
      		 10,056,684 9,167,183 0.91×
      switch
      		 87,998 20,674,719 234.94×
      timermix
      		 208,833 15,237,330 72.96×
      tlb-shootdown
      		 8,666 110,213 12.72×
      tsc
      		 747,548 52,776,779 70.60×
      usersyscall
      		 623,154 13,653,135 21.91×
      vdso
      		 106,566,245 5,872,786,030 55.11×
      vfork
      		 28,055 119,614 4.26×
      x86syscall
      		 8,326,637 547,460,850 65.75×
      yield
      		 80,035,390 87,598,378 1.09×
      zero
      		 4,355,417 239,827,196 55.06×
      CPU float/SIMDSingleMultiScaling
      eigen
      		 13,891 828,716 59.66×
      expmath
      		 2,804 189,834 67.70×
      fma
      		 621,429 51,498,779 82.87×
      fp
      		 190.81 9,231 48.38×
      fractal
      		 4.8 332.06 69.18×
      hyperbolic
      		 2,991 203,622 68.07×
      logmath
      		 4,714 316,662 67.17×
      matrix
      		 3,446 225,120 65.33×
      matrix-3d
      		 601.2 11,041 18.37×
      monte-carlo
      		 247.27 15,452 62.49×
      mpfr
      		 1,373 65,729 47.89×
      powmath
      		 1,641 111,520 67.96×
      trig
      		 1,915 108,039 56.41×
      veccmp
      		 19,832 864,172 43.57×
      vecfp
      		 1,484 85,434 57.55×
      vecmath
      		 3,150 204,880 65.04×
      vecshuf
      		 646.26 32,797 50.75×
      vecwide
      		 31,055 1,525,198 49.11×
      AlgorithmsSingleMultiScaling
      bitonicsort
      		 7.8 383.99 49.23×
      bsearch
      		 240.49 15,704 65.30×
      bubblesort
      		 1 38.4 38.40×
      fibsearch
      		 280.34 16,504 58.87×
      heapsort
      		 9.2 479.98 52.17×
      hsearch
      		 7,117 354,187 49.77×
      insertionsort
      		 10.2 287.8 28.22×
      malloc
      		 1,822,255 67,106,481 36.83×
      mcontend
      		 65.12 16,632 255.40×
      mergesort
      		 26 1,580 60.76×
      qsort
      		 20 951.37 47.57×
      skiplist
      		 5,283 357,098 67.59×
      tree
      		 1 57.17 57.17×
      MemorySingleMultiScaling
      cache
      		 247,136 2,313,631 9.36×
      cachehammer
      		 351,697 13,501,795 38.39×
      cacheline
      		 35.17 3,531 100.40×
      l1cache
      		 892.67 16,910 18.94×
      memcpy
      		 393.76 17,672 44.88×
      memrate
      		 1.58 33.63 21.28×
      memthrash
      		 300.14 1,300 4.33×
      ptr-chase
      		 10,388,594 756,113,992 72.78×
      spinmem
      		 218.19 4,295 19.68×
      stream
      		 15.56 25,911 1665.25×
      vm
      		 102,668 308,028 3.00×

      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
      n2-highcpu-2 22 GiB0
      n2-standard-2 28 GiB0
      n2-highmem-2 216 GiB0
      n2-highcpu-4 44 GiB0
      n2-standard-4 416 GiB0
      n2-highmem-4 432 GiB0
      n2-highcpu-8 88 GiB0

      Similar Servers

      INSTANCEVENDORvCPUsMEMORYAccelerators

      n2-highcpu-96 FAQs