n1-highcpu-96 by Google Cloud Platform

n1-highcpu-96 is a 96 vCPUs, 86 GB RAM server offered by Google Cloud Platform with 96 vCPUs, 86 GiB of memory and 0 GB of storage. The pricing starts at 0.5476 USD per hour.
96 vCPU
86 GiB Memory
Spare SCore
70,482
(All-cores)
1,117
(Single-core)

Specifications

Server Metadata

Vendor ID
gcp
Server ID
4096
Name
n1-highcpu-96
Description
96 vCPUs, 86 GB RAM
Family
n1
Hw Virt
Average Time To Start
23
Status
active
Observed At
2026-10-21T10:42:09.465560

Availability

REGION / IDSPOTONDEMAND
Council Bluffs (US) / us-central1
1.7482 USD/h 3.4007 USD/h
The Dalles (US) / us-west1
1.7482 USD/h 3.4007 USD/h
Moncks Corner (US) / us-east1
1.7482 USD/h 3.4007 USD/h
Tel Aviv (IL) / me-west1
2.057 USD/h 3.7408 USD/h
St. Ghislain (BE) / europe-west1
1.9246 USD/h 3.7409 USD/h
Toronto (CA) / northamerica-northeast2
1.7203 USD/h 3.744 USD/h
Montréal (CA) / northamerica-northeast1
2.2311 USD/h 3.744 USD/h
Eemshaven (NL) / europe-west4
2.2408 USD/h 3.744 USD/h
Hamina (FI) / europe-north1
0.808 USD/h 3.7443 USD/h
Las Vegas (US) / us-west4
1.5644 USD/h 3.8303 USD/h
Ashburn (US) / us-east4
1.6311 USD/h 3.8303 USD/h
Changhua County (TW) / asia-east1
0.839 USD/h 3.9377 USD/h
Delhi (IN) / asia-south2
1.0169 USD/h 4.0847 USD/h
Mumbai (IN) / asia-south1
1.4458 USD/h 4.0847 USD/h
Salt Lake City (US) / us-west3
1.4795 USD/h 4.0848 USD/h
Los Angeles (US) / us-west2
1.8738 USD/h 4.0848 USD/h
Warsaw (PL) / europe-central2
1.3444 USD/h 4.115 USD/h
Jurong West (SG) / asia-southeast1
1.0028 USD/h 4.1954 USD/h
Tokyo (JP) / asia-northeast1
1.656 USD/h 4.3675 USD/h
Osaka (JP) / asia-northeast2
2.0173 USD/h 4.3675 USD/h
Seoul (KR) / asia-northeast3
2.1344 USD/h 4.3675 USD/h
London (GB) / europe-west2
1.6438 USD/h 4.3817 USD/h
Frankfurt (DE) / europe-west3
2.2344 USD/h 4.3817 USD/h
Zurich (CH) / europe-west6
1.8579 USD/h 4.4548 USD/h
Jakarta (ID) / asia-southeast2
0.5476 USD/h 4.573 USD/h
Hong Kong / asia-east2
0.7412 USD/h 4.7584 USD/h
Melbourne (AU) / australia-southeast2
0.9059 USD/h 4.8255 USD/h
Sydney (AU) / australia-southeast1
2.3689 USD/h 4.8255 USD/h
Osasco (BR) / southamerica-east1
1.7997 USD/h 5.3983 USD/h

Processor

vCPUs
96
CPU Allocation
Dedicated
CPU Cores
48
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU Family
Xeon
CPU L1D Cache
32 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
48 MiB
CPU L3 Cache
39 MiB
CPU L3 Cache Total
77 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, pti, ssbd, ibrs, ibpb, stibp, fsgsbase, tsc_adjust, bmi1, hle, avx2, smep, bmi2, erms, invpcid, rtm, mpx, avx512f, avx512dq, rdseed, adx, smap, clflushopt, clwb, avx512cd, avx512bw, avx512vl, xsaveopt, xsavec, xgetbv1, xsaves, arat, md_clear, arch_capabilities
Ecpus
63.1
Scalability
65.73

System Resources and Accelerators

MEMORY
Memory Amount
86 GiB
Memory Amount Actual
86 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-optimized virtual machine designed for highly parallelized workloads, multi-threaded processing, and high-throughput memory read operations.

      Compute OptimizedGeneral Purpose

      Google Cloud Platform n1-highcpu-96 is a compute-optimized virtual machine instance featuring 96 vCPUs and 86.0 GB of RAM. Powered by dedicated Intel Xeon x86_64 processors with 48 physical cores and Hyper-Threading, this instance is designed for compute-heavy, parallel workloads. It does not include bundled local storage or GPUs. Benchmarks indicate weak single-core performance but strong multi-core execution, placing in the top 25% for stress-ng multi-core tests. Memory read bandwidth is also highly efficient, scoring in the top quartile. Qualitatively, the instance provides high compute density for parallel processing but is constrained by a low memory-to-vCPU ratio of 0.9 GB per core. It is best suited for highly parallelized workloads such as software compilation, multi-threaded data processing, Redis operations, and static web serving.

      Economics

      Average Price per Region

      Prices per Zone

      Lowest Prices

      Performance

      Workload Profiles

      2.3Score
      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) 6,033,384 ops/sec 1,918,343 ops/sec 3.2 50.00% 50.00% +79.00%
      Redis RPS (pipeline=16, SET) 38,689,635 ops/sec 11,704,734 ops/sec 3.2 20.00% 20.00% +26.00%
      PassMark Memory Mark (composite) 2,524 2,416 1 10.00% 10.00% 0.00%
      Memory bandwidth (read, 16 MB ~ L3) 201,387 MB/sec 108,420 MB/sec 1.8 10.00% 10.00% +6.10%
      PassMark single-thread CPU 1,580 Mops/s 2,365 Mops/s 0.67 10.00% 10.00% -3.90%

      Memory Bandwidth

      Compression

      OpenSSL

      Geekbench Single-Core

      Score: 1,057

      Geekbench Multi-Core

      Score: 10,778

      Passmark CPU Scores

      BENCHMARKSCORE
      Mark
      		 46,528
      Compression
      		 996,212
      Encryption
      		 30,618
      Extended Instructions
      		 63,695
      Floating Point Maths
      		 124,674
      Integer Maths
      		 232,267
      Physics
      		 4,799
      Prime Numbers
      		 309
      Single Threaded
      		 1,580
      String Sorting
      		 113,471

      Passmark Memory Scores

      BENCHMARKSCORE
      Memory Mark
      		 2,524
      Database Operations
      		 19,544
      Memory Latency
      		 53
      Memory Read Cached
      		 19,358
      Memory Read Uncached
      		 10,250
      Memory Write
      		 9,696

      Stress-ng Stressors

      CPU integerSingleMultiScaling
      atomic
      		 221.93 103.53 0.47×
      bitops
      		 102,722 4,890,886 47.61×
      branch
      		 140,862 8,860,700 62.90×
      funccall
      		 495.26 21,222 42.85×
      funcret
      		 10,366 612,258 59.07×
      hash
      		 115,603 6,829,524 59.08×
      intmath
      		 20,879 1,261,605 60.42×
      nop
      		 174,558 6,267,731 35.91×
      prime
      		 289,354 13,574,430 46.91×
      rotate
      		 88,174 5,040,633 57.17×
      str
      		 661,306 30,082,773 45.49×
      wcs
      		 433,347 18,380,507 42.42×
      CryptographySingleMultiScaling
      crypt
      		 2,219 90,121 40.61×
      CompressionSingleMultiScaling
      jpeg
      		 511.66 25,604 50.04×
      zlib
      		 47.51 2,972 62.56×
      HypervisorSingleMultiScaling
      clock
      		 14,070 161,795 11.50×
      clone
      		 256.99 8,137 31.66×
      context
      		 1,132 60,644 53.56×
      cpu-sched
      		 56.54 6,229 110.17×
      eventfd
      		 39,326 1,386,769 35.26×
      fault
      		 8,364 136,697 16.34×
      fork
      		 2,455 54,500 22.20×
      getrandom
      		 38,641 2,005,028 51.89×
      hrtimers
      		 349,667 6,839,185 19.56×
      mprotect
      		 575,746 7,734,105 13.43×
      mtx
      		 13,689,441 1,710,144,469 124.92×
      munmap
      		 1,721 57,258 33.28×
      null
      		 111,399 5,719,569 51.34×
      pipe
      		 693,769 13,773,551 19.85×
      priv-instr
      		 286,929 8,524,927 29.71×
      resched
      		 3,330,234 3,588,121 1.08×
      switch
      		 86,600 8,768,606 101.25×
      timermix
      		 140,670 7,385,082 52.50×
      tlb-shootdown
      		 6,525 70,488 10.80×
      tsc
      		 901,454 52,618,663 58.37×
      usersyscall
      		 187,211 7,997,496 42.72×
      vdso
      		 75,347,737 4,383,265,092 58.17×
      vfork
      		 15,910 123,060 7.73×
      x86syscall
      		 1,255,555 64,434,901 51.32×
      yield
      		 27,302,424 28,854,988 1.06×
      zero
      		 958,005 50,241,861 52.44×
      CPU float/SIMDSingleMultiScaling
      eigen
      		 14,765 720,279 48.78×
      expmath
      		 2,442 154,776 63.39×
      fma
      		 525,075 34,197,753 65.13×
      fp
      		 147.98 8,066 54.51×
      fractal
      		 4.8 290.44 60.51×
      hyperbolic
      		 2,552 165,524 64.86×
      logmath
      		 4,076 253,401 62.16×
      matrix
      		 2,773 174,848 63.05×
      matrix-3d
      		 608.9 6,028 9.90×
      monte-carlo
      		 64.05 3,796 59.27×
      mpfr
      		 1,102 49,050 44.49×
      powmath
      		 1,439 91,507 63.59×
      trig
      		 1,526 84,185 55.17×
      veccmp
      		 13,979 579,435 41.45×
      vecfp
      		 2,014 91,658 45.52×
      vecmath
      		 1,914 126,427 66.06×
      vecshuf
      		 810.3 34,131 42.12×
      vecwide
      		 49,840 2,245,033 45.04×
      AlgorithmsSingleMultiScaling
      bitonicsort
      		 6.4 308.59 48.22×
      bsearch
      		 220.95 10,825 48.99×
      bubblesort
      		 0.6 19.2 32.00×
      fibsearch
      		 220.98 10,656 48.22×
      heapsort
      		 7.8 386.58 49.56×
      hsearch
      		 4,559 233,164 51.14×
      insertionsort
      		 7.4 307.2 41.51×
      malloc
      		 1,985,573 54,678,698 27.54×
      mcontend
      		 69.87 11,290 161.59×
      mergesort
      		 21.2 1,137 53.61×
      qsort
      		 13.4 636.78 47.52×
      skiplist
      		 4,395 259,769 59.11×
      tree
      		 1.4 38.38 27.41×
      MemorySingleMultiScaling
      cache
      		 284,878 1,685,045 5.91×
      cachehammer
      		 340,052 12,077,554 35.52×
      cacheline
      		 15.76 1,390 88.19×
      l1cache
      		 204.35 8,776 42.95×
      memcpy
      		 316.99 14,039 44.29×
      memrate
      		 1.16 14.72 12.69×
      memthrash
      		 247.14 1,014 4.10×
      ptr-chase
      		 29,667,815 778,967,507 26.26×
      spinmem
      		 196.59 3,244 16.50×
      stream
      		 13.36 21,739 1627.17×
      vm
      		 33,985 309,470 9.11×

      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
      n1-standard-1 14 GiB0
      n1-highcpu-2 22 GiB0
      n1-standard-2 28 GiB0
      n1-highmem-2 213 GiB0
      n1-highcpu-4 44 GiB0
      n1-standard-4 415 GiB0
      n1-highmem-4 426 GiB0

      Similar Servers

      INSTANCEVENDORvCPUsMEMORYAccelerators

      n1-highcpu-96 FAQs