n4d-standard-96 by Google Cloud Platform

n4d-standard-96 is a 96 vCPUs, 384 GB RAM server offered by Google Cloud Platform with 96 vCPUs, 384 GiB of memory and 0 GB of storage. The pricing starts at 0.7889 USD per hour.
96 vCPU
384 GiB Memory
Spare SCore
165,170
(All-cores)
3,437
(Single-core)

Specifications

Server Metadata

Vendor ID
gcp
Server ID
1751008
Name
n4d-standard-96
Description
96 vCPUs, 384 GB RAM
Family
n4d
Hw Virt
Average Time To Start
18
Status
active
Observed At
2026-10-20T03:53:47.784852

Availability

REGION / IDSPOTONDEMAND
Mumbai (IN) / asia-south1
1.5602 USD/h 3.0916 USD/h
Moncks Corner (US) / us-east1
1.1964 USD/h 4.0656 USD/h
The Dalles (US) / us-west1
1.2524 USD/h 4.0656 USD/h
Columbus (US) / us-east5
2.1646 USD/h 4.0656 USD/h
Ashburn (US) / us-east4
2.2733 USD/h 4.0656 USD/h
Council Bluffs (US) / us-central1
2.4378 USD/h 4.0656 USD/h
Delhi (IN) / asia-south2
1.3811 USD/h 4.2273 USD/h
Eemshaven (NL) / europe-west4
2.5617 USD/h 4.2707 USD/h
Tel Aviv (IL) / me-west1
0.7889 USD/h 4.4718 USD/h
St. Ghislain (BE) / europe-west1
1.3428 USD/h 4.4748 USD/h
Hamina (FI) / europe-north1
1.8518 USD/h 4.4748 USD/h
Las Vegas (US) / us-west4
1.2363 USD/h 4.5799 USD/h
London (GB) / europe-west2
2.779 USD/h 4.6334 USD/h
Frankfurt (DE) / europe-west3
2.5523 USD/h 4.799 USD/h
Salt Lake City (US) / us-west3
2.9307 USD/h 4.8876 USD/h
Jurong West (SG) / asia-southeast1
2.8067 USD/h 5.0144 USD/h
Dammam (SA) / me-central2
1.7295 USD/h 6.5065 USD/h

Processor

vCPUs
96
CPU Allocation
Dedicated
CPU Cores
48
CPU Architecture
x86_64
CPU Manufacturer
AMD
CPU Family
EPYC
CPU Model
9B45
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
48 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, xtopology, 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, ibrs_enhanced, vmmcall, fsgsbase, tsc_adjust, bmi1, avx2, smep, bmi2, invpcid, avx512f, avx512dq, rdseed, adx, smap, avx512ifma, clflushopt, clwb, avx512cd, sha_ni, avx512bw, avx512vl, xsaveopt, xsavec, xgetbv1, xsaves, avx_vnni, avx512_bf16, clzero, xsaveerptr, wbnoinvd, arat, avx512vbmi, umip, pku, ospke, avx512_vbmi2, gfni, vaes, vpclmulqdq, avx512_vnni, avx512_bitalg, avx512_vpopcntdq, rdpid, movdiri, movdir64b, avx512_vp2intersect, flush_l1d
Ecpus
48.1
Scalability
50.1

System Resources and Accelerators

MEMORY
Memory Amount
384 GiB
Memory Amount Actual
384 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, 96-vCPU AMD EPYC platform delivering top-tier compute performance and balanced memory for demanding enterprise workloads.

      General PurposeCompute Optimized

      Google Cloud Platform n4d-standard-96 is a general-purpose server featuring 96 dedicated vCPUs and 384 GB of RAM, powered by an AMD EPYC 9B45 processor on the x86_64 architecture. With 48 physical cores and multithreading, it delivers top-tier CPU performance in both single-core and multi-core benchmarks. The hardware includes advanced instruction sets such as AVX-512 and AES, which drive exceptional throughput in cryptographic, database, and vector operations. Although it lacks local storage and GPU accelerators, its balanced 4 GB of memory per vCPU and high memory bandwidth make it highly efficient for demanding workloads. It is well-suited for software compilation, large-scale web serving, Redis databases, and CPU-based machine learning inference.

      Economics

      Average Price per Region

      Prices per Zone

      Lowest Prices

      Performance

      Workload Profiles

      6.5Score
      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) 21,683,449 ops/sec 1,917,050 ops/sec 12 50.00% 50.00% +250.00%
      Redis RPS (pipeline=16, SET) 116,786,949 ops/sec 11,661,818 ops/sec 10 20.00% 20.00% +58.00%
      PassMark Memory Mark (composite) 2,968 2,417 1.2 10.00% 10.00% +1.80%
      Memory bandwidth (read, 16 MB ~ L3) 306,386 MB/sec 108,361 MB/sec 2.8 10.00% 10.00% +11.00%
      PassMark single-thread CPU 3,463 Mops/s 2,365 Mops/s 1.5 10.00% 10.00% +4.10%

      Memory Bandwidth

      Compression

      OpenSSL

      Geekbench Single-Core

      Score: 2,590

      Geekbench Multi-Core

      Score: 27,376

      Passmark CPU Scores

      BENCHMARKSCORE
      Mark
      		 115,230
      Compression
      		 2,139,828
      Encryption
      		 116,573
      Extended Instructions
      		 165,643
      Floating Point Maths
      		 355,665
      Integer Maths
      		 583,093
      Physics
      		 17,665
      Prime Numbers
      		 839
      Single Threaded
      		 3,463
      String Sorting
      		 258,361

      Passmark Memory Scores

      BENCHMARKSCORE
      Memory Mark
      		 2,968
      Database Operations
      		 40,030
      Memory Latency
      		 76
      Memory Read Cached
      		 31,311
      Memory Read Uncached
      		 29,310
      Memory Write
      		 23,027

      Stress-ng Stressors

      CPU integerSingleMultiScaling
      atomic
      		 529.56 248.35 0.47×
      bitops
      		 210,613 11,585,534 55.01×
      branch
      		 167,931 13,182,867 78.50×
      funccall
      		 827.96 50,640 61.16×
      funcret
      		 24,668 1,653,981 67.05×
      hash
      		 199,601 14,655,509 73.42×
      intmath
      		 73,650 4,188,991 56.88×
      nop
      		 392,342 22,389,667 57.07×
      prime
      		 570,720 34,461,438 60.38×
      rotate
      		 130,044 6,344,682 48.79×
      str
      		 1,476,288 78,993,302 53.51×
      wcs
      		 980,322 48,096,791 49.06×
      CryptographySingleMultiScaling
      crypt
      		 4,816 256,274 53.21×
      CompressionSingleMultiScaling
      jpeg
      		 1,155 69,023 59.76×
      zlib
      		 82.05 5,639 68.73×
      HypervisorSingleMultiScaling
      clock
      		 43,747 326,479 7.46×
      clone
      		 353.49 8,757 24.77×
      context
      		 8,409 680,112 80.88×
      cpu-sched
      		 56.52 6,068 107.35×
      eventfd
      		 68,228 4,706,056 68.98×
      fault
      		 39,679 229,808 5.79×
      fork
      		 6,286 83,685 13.31×
      getrandom
      		 89,712 4,619,322 51.49×
      hrtimers
      		 740,584 1,259,858 1.70×
      mprotect
      		 740,566 26,038,286 35.16×
      mtx
      		 49,529,540 5,717,931,624 115.44×
      munmap
      		 4,946 106,707 21.58×
      null
      		 1,380,105 99,598,067 72.17×
      pipe
      		 2,530,967 58,835,661 23.25×
      priv-instr
      		 1,277,987 33,030,668 25.85×
      resched
      		 14,868,836 16,521,928 1.11×
      switch
      		 128,859 10,820,155 83.97×
      timermix
      		 512,415 5,887,478 11.49×
      tlb-shootdown
      		 19,091 246,095 12.89×
      tsc
      		 810,470 74,171,593 91.52×
      usersyscall
      		 1,058,116 32,873,919 31.07×
      vdso
      		 99,285,332 8,337,761,980 83.98×
      vfork
      		 40,523 200,997 4.96×
      x86syscall
      		 18,082,480 1,348,322,269 74.57×
      yield
      		 137,034,736 153,949,974 1.12×
      zero
      		 7,177,099 451,145,388 62.86×
      CPU float/SIMDSingleMultiScaling
      eigen
      		 16,005 894,732 55.90×
      expmath
      		 3,669 227,813 62.10×
      fma
      		 1,434,789 96,414,320 67.20×
      fp
      		 177.06 14,320 80.88×
      fractal
      		 8.2 573.29 69.91×
      hyperbolic
      		 5,023 339,754 67.64×
      logmath
      		 5,218 381,383 73.09×
      matrix
      		 8,518 332,255 39.01×
      matrix-3d
      		 1,405 12,297 8.75×
      monte-carlo
      		 421.06 31,586 75.02×
      mpfr
      		 2,212 138,956 62.82×
      powmath
      		 1,691 116,685 69.01×
      trig
      		 2,236 153,340 68.59×
      veccmp
      		 27,560 1,740,140 63.14×
      vecfp
      		 2,244 160,537 71.54×
      vecmath
      		 4,368 317,247 72.63×
      vecshuf
      		 921.94 53,109 57.61×
      vecwide
      		 36,892 2,239,670 60.71×
      AlgorithmsSingleMultiScaling
      bitonicsort
      		 11.6 644.58 55.57×
      bsearch
      		 335.75 22,574 67.23×
      bubblesort
      		 1 60 60.00×
      fibsearch
      		 395.62 25,781 65.17×
      heapsort
      		 13 843.38 64.88×
      hsearch
      		 12,234 659,622 53.92×
      insertionsort
      		 14.8 527.8 35.66×
      malloc
      		 4,815,155 171,820,861 35.68×
      mcontend
      		 132.1 16,576 125.48×
      mergesort
      		 35.4 2,411 68.12×
      qsort
      		 23.8 1,610 67.65×
      skiplist
      		 7,659 548,879 71.66×
      tree
      		 2.4 85.8 35.75×
      MemorySingleMultiScaling
      cache
      		 1,051,741 1,840,612 1.75×
      cachehammer
      		 283,993 17,485,034 61.57×
      cacheline
      		 39.38 4,621 117.34×
      l1cache
      		 1,254 35,791 28.54×
      memcpy
      		 743.57 34,387 46.25×
      memrate
      		 0.96 17.77 18.51×
      memthrash
      		 334.6 2,734 8.17×
      ptr-chase
      		 73,124,571 727,015,395 9.94×
      spinmem
      		 363.19 7,143 19.67×
      stream
      		 147.44 130,187 882.98×
      vm
      		 232,506 304,780 1.31×

      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
      n4d-highcpu-2 24 GiB0
      n4d-standard-2 28 GiB0
      n4d-highmem-2 216 GiB0
      n4d-highcpu-4 48 GiB0
      n4d-standard-4 416 GiB0
      n4d-highmem-4 432 GiB0
      n4d-highcpu-8 816 GiB0

      Similar Servers

      INSTANCEVENDORvCPUsMEMORYAccelerators

      n4d-standard-96 FAQs