n4d-highmem-96 by Google Cloud Platform

n4d-highmem-96 is a 96 vCPUs, 768 GB RAM server offered by Google Cloud Platform with 96 vCPUs, 768 GiB of memory and 0 GB of storage. The pricing starts at 1.1393 USD per hour.
96 vCPU
768 GiB Memory
Spare SCore
164,007
(All-cores)
3,423
(Single-core)

Specifications

Server Metadata

Vendor ID
gcp
Server ID
1751108
Name
n4d-highmem-96
Description
96 vCPUs, 768 GB RAM
Family
n4d
Hw Virt
Average Time To Start
21
Status
active
Observed At
2026-10-28T08:13:57.627546

Availability

REGION / IDSPOTONDEMAND
Mumbai (IN) / asia-south1
2.049 USD/h 4.0593 USD/h
Moncks Corner (US) / us-east1
1.4911 USD/h 5.3366 USD/h
The Dalles (US) / us-west1
1.5613 USD/h 5.3366 USD/h
Columbus (US) / us-east5
2.8412 USD/h 5.3366 USD/h
Ashburn (US) / us-east4
2.9837 USD/h 5.3366 USD/h
Council Bluffs (US) / us-central1
3.2004 USD/h 5.3366 USD/h
Delhi (IN) / asia-south2
1.7226 USD/h 5.5482 USD/h
Eemshaven (NL) / europe-west4
3.3627 USD/h 5.607 USD/h
Tel Aviv (IL) / me-west1
1.1393 USD/h 5.8695 USD/h
St. Ghislain (BE) / europe-west1
1.7622 USD/h 5.8726 USD/h
Hamina (FI) / europe-north1
2.3088 USD/h 5.8726 USD/h
Las Vegas (US) / us-west4
1.623 USD/h 6.0122 USD/h
London (GB) / europe-west2
3.6476 USD/h 6.0811 USD/h
Frankfurt (DE) / europe-west3
3.3506 USD/h 6.3005 USD/h
Salt Lake City (US) / us-west3
3.8473 USD/h 6.416 USD/h
Jurong West (SG) / asia-southeast1
3.6837 USD/h 6.5811 USD/h
Dammam (SA) / me-central2
2.1573 USD/h 8.5417 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
47.9
Scalability
49.9

System Resources and Accelerators

MEMORY
Memory Amount
768 GiB
Memory Amount Actual
768 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

      An x86_64 server featuring ninety-six vCPUs and 768 GB RAM for memory-intensive databases and multi-threaded enterprise workloads.

      Memory OptimizedGeneral Purpose

      Google Cloud Platform n4d-highmem-96 is an x86_64 memory-optimized server featuring 96 vCPUs and 768 GB of RAM, powered by a dedicated AMD EPYC 9B45 processor. With 48 physical cores and a 1:8 vCPU-to-RAM ratio, it delivers top-tier CPU and memory performance. Benchmarks show top 10% results in multi-core processing, Redis operations, and static web serving, alongside strong LLM inference capabilities. However, the server lacks local storage and GPU accelerators, requiring network-attached storage. It is designed for memory-intensive databases, software compilation, and heavy multi-threaded workloads.

      Economics

      Average Price per Region

      Prices per Zone

      Lowest Prices

      Performance

      Workload Profiles

      6Score
      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) 19,732,157 ops/sec 1,919,289 ops/sec 11 50.00% 50.00% +230.00%
      Redis RPS (pipeline=16, SET) 105,355,522 ops/sec 11,728,058 ops/sec 9.2 20.00% 20.00% +56.00%
      PassMark Memory Mark (composite) 2,900 2,416 1.2 10.00% 10.00% +1.80%
      Memory bandwidth (read, 16 MB ~ L3) 312,024 MB/sec 108,466 MB/sec 2.8 10.00% 10.00% +11.00%
      PassMark single-thread CPU 3,419 Mops/s 2,365 Mops/s 1.4 10.00% 10.00% +3.40%

      Memory Bandwidth

      Compression

      OpenSSL

      Geekbench Single-Core

      Score: 2,542

      Geekbench Multi-Core

      Score: 26,419

      Passmark CPU Scores

      BENCHMARKSCORE
      Mark
      		 111,595
      Compression
      		 1,961,732
      Encryption
      		 111,910
      Extended Instructions
      		 151,166
      Floating Point Maths
      		 343,606
      Integer Maths
      		 575,609
      Physics
      		 17,612
      Prime Numbers
      		 850
      Single Threaded
      		 3,419
      String Sorting
      		 243,836

      Passmark Memory Scores

      BENCHMARKSCORE
      Memory Mark
      		 2,900
      Database Operations
      		 37,266
      Memory Latency
      		 78
      Memory Read Cached
      		 31,105
      Memory Read Uncached
      		 28,679
      Memory Write
      		 22,531

      Stress-ng Stressors

      CPU integerSingleMultiScaling
      atomic
      		 523.34 241.43 0.46×
      bitops
      		 206,265 11,342,947 54.99×
      branch
      		 156,003 12,888,446 82.62×
      funccall
      		 821.4 48,552 59.11×
      funcret
      		 23,889 1,528,425 63.98×
      hash
      		 198,696 14,136,536 71.15×
      intmath
      		 72,967 4,170,779 57.16×
      nop
      		 383,419 22,052,832 57.52×
      prime
      		 566,133 33,042,299 58.36×
      rotate
      		 126,098 6,249,012 49.56×
      str
      		 1,456,511 74,001,059 50.81×
      wcs
      		 942,788 45,491,875 48.25×
      CryptographySingleMultiScaling
      crypt
      		 4,960 225,087 45.38×
      CompressionSingleMultiScaling
      jpeg
      		 1,153 55,437 48.09×
      zlib
      		 67.93 5,561 81.87×
      HypervisorSingleMultiScaling
      clock
      		 43,589 317,578 7.29×
      clone
      		 349.26 8,691 24.88×
      context
      		 8,371 663,764 79.30×
      cpu-sched
      		 64.23 6,181 96.23×
      eventfd
      		 67,727 4,104,732 60.61×
      fault
      		 39,018 236,949 6.07×
      fork
      		 6,049 73,869 12.21×
      getrandom
      		 89,706 4,440,824 49.50×
      hrtimers
      		 134,504 1,195,384 8.89×
      mprotect
      		 697,922 24,455,058 35.04×
      mtx
      		 49,582,126 5,021,524,892 101.28×
      munmap
      		 4,843 96,624 19.95×
      null
      		 1,352,834 100,891,311 74.58×
      pipe
      		 2,524,625 59,140,671 23.43×
      priv-instr
      		 1,322,909 32,483,918 24.55×
      resched
      		 15,259,348 16,922,279 1.11×
      switch
      		 127,202 38,622,522 303.63×
      timermix
      		 509,570 5,796,316 11.37×
      tlb-shootdown
      		 20,867 246,682 11.82×
      tsc
      		 798,824 75,672,169 94.73×
      usersyscall
      		 1,025,736 30,682,947 29.91×
      vdso
      		 96,808,023 7,888,754,025 81.49×
      vfork
      		 38,854 178,616 4.60×
      x86syscall
      		 17,569,212 1,301,077,438 74.05×
      yield
      		 124,162,921 146,742,882 1.18×
      zero
      		 7,104,483 413,091,772 58.15×
      CPU float/SIMDSingleMultiScaling
      eigen
      		 15,753 829,186 52.64×
      expmath
      		 3,664 200,311 54.67×
      fma
      		 1,421,961 89,852,200 63.19×
      fp
      		 190.95 13,594 71.19×
      fractal
      		 8 534.18 66.77×
      hyperbolic
      		 4,974 349,670 70.30×
      logmath
      		 5,137 345,439 67.25×
      matrix
      		 8,339 282,349 33.86×
      matrix-3d
      		 1,390 13,257 9.53×
      monte-carlo
      		 420.78 27,283 64.84×
      mpfr
      		 2,200 125,839 57.20×
      powmath
      		 1,669 118,100 70.75×
      trig
      		 2,232 137,416 61.56×
      veccmp
      		 27,490 1,562,497 56.84×
      vecfp
      		 2,188 156,113 71.36×
      vecmath
      		 4,256 295,871 69.52×
      vecshuf
      		 899.47 46,540 51.74×
      vecwide
      		 36,383 2,010,505 55.26×
      AlgorithmsSingleMultiScaling
      bitonicsort
      		 11.4 574 50.35×
      bsearch
      		 331.77 21,239 64.02×
      bubblesort
      		 1.2 57.8 48.17×
      fibsearch
      		 386.5 25,753 66.63×
      heapsort
      		 13 756.58 58.20×
      hsearch
      		 11,969 596,295 49.82×
      insertionsort
      		 14.6 478.8 32.79×
      malloc
      		 4,656,605 162,902,434 34.98×
      mcontend
      		 118.09 16,339 138.36×
      mergesort
      		 35.4 2,150 60.73×
      qsort
      		 23 1,417 61.61×
      skiplist
      		 7,500 534,782 71.30×
      tree
      		 2.2 78.96 35.89×
      MemorySingleMultiScaling
      cache
      		 1,023,452 1,857,295 1.81×
      cachehammer
      		 290,547 16,547,088 56.95×
      cacheline
      		 52.18 4,470 85.67×
      l1cache
      		 1,251 32,300 25.81×
      memcpy
      		 719.42 30,021 41.73×
      memrate
      		 1.19 21.52 18.08×
      memthrash
      		 343.89 2,709 7.88×
      ptr-chase
      		 71,215,699 753,223,520 10.58×
      spinmem
      		 350.78 7,379 21.04×
      stream
      		 144.73 146,100 1009.47×
      vm
      		 232,447 304,591 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-highmem-96 FAQs