n2d-highmem-96 by Google Cloud Platform

n2d-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 1.5468 USD per hour.
96 vCPU
768 GiB Memory
Spare SCore
129,682
(All-cores)
2,925
(Single-core)

Specifications

Server Metadata

Vendor ID
gcp
Server ID
912096
Name
n2d-highmem-96
Description
96 vCPUs 768 GB RAM
Family
n2d
Hw Virt
Average Time To Start
17
Status
active
Observed At
2026-10-25T01:49:54.626199

Availability

REGION / IDSPOTONDEMAND
Mumbai (IN) / asia-south1
2.1608 USD/h 3.611 USD/h
Columbus (US) / us-east5
1.7639 USD/h 5.471 USD/h
Council Bluffs (US) / us-central1
2.5137 USD/h 5.471 USD/h
The Dalles (US) / us-west1
2.5137 USD/h 5.471 USD/h
Moncks Corner (US) / us-east1
2.5137 USD/h 5.471 USD/h
Tel Aviv (IL) / me-west1
1.7892 USD/h 6.0181 USD/h
Johannesburg (ZA) / africa-south1
3.168 USD/h 6.0185 USD/h
St. Ghislain (BE) / europe-west1
3.6109 USD/h 6.0185 USD/h
Toronto (CA) / northamerica-northeast2
3.0966 USD/h 6.0232 USD/h
Eemshaven (NL) / europe-west4
3.0985 USD/h 6.0232 USD/h
Montréal (CA) / northamerica-northeast1
3.5864 USD/h 6.0232 USD/h
Hamina (FI) / europe-north1
3.4581 USD/h 6.0235 USD/h
Las Vegas (US) / us-west4
1.8234 USD/h 6.1617 USD/h
Ashburn (US) / us-east4
3.504 USD/h 6.1617 USD/h
Changhua County (TW) / asia-east1
3.2582 USD/h 6.3348 USD/h
Milan (IT) / europe-west8
1.5468 USD/h 6.3464 USD/h
Paris (FR) / europe-west9
2.6172 USD/h 6.3464 USD/h
Dallas (US) / us-south1
2.7016 USD/h 6.4558 USD/h
Madrid (ES) / europe-southwest1
3.8736 USD/h 6.4558 USD/h
Salt Lake City (US) / us-west3
2.7483 USD/h 6.5712 USD/h
Los Angeles (US) / us-west2
3.7265 USD/h 6.5712 USD/h
Delhi (IN) / asia-south2
3.7265 USD/h 6.5712 USD/h
Melbourne (AU) / australia-southeast2
1.8564 USD/h 6.5989 USD/h
Warsaw (PL) / europe-central2
3.9612 USD/h 6.6199 USD/h
Doha (QA) / me-central1
2.7423 USD/h 6.6473 USD/h
Jurong West (SG) / asia-southeast1
1.5878 USD/h 6.7493 USD/h
Tokyo (JP) / asia-northeast1
4.1967 USD/h 7.0136 USD/h
Osaka (JP) / asia-northeast2
1.5665 USD/h 7.0262 USD/h
Seoul (KR) / asia-northeast3
1.7255 USD/h 7.0262 USD/h
Berlin (DE) / europe-west10
2.0845 USD/h 7.0482 USD/h
Frankfurt (DE) / europe-west3
3.6169 USD/h 7.0482 USD/h
London (GB) / europe-west2
4.229 USD/h 7.0482 USD/h
Turin (IT) / europe-west12
1.5711 USD/h 7.0576 USD/h
Zurich (CH) / europe-west6
3.6858 USD/h 7.1672 USD/h
Jakarta (ID) / asia-southeast2
2.8585 USD/h 7.3567 USD/h
Hong Kong / asia-east2
3.6639 USD/h 7.6548 USD/h
Sydney (AU) / australia-southeast1
4.6464 USD/h 7.763 USD/h
Santiago (CL) / southamerica-west1
2.102 USD/h 7.8236 USD/h
Osasco (BR) / southamerica-east1
2.5868 USD/h 8.6846 USD/h
Dammam (SA) / me-central2
1.8724 USD/h 8.7537 USD/h

Processor

vCPUs
96
CPU Allocation
Dedicated
CPU Cores
48
CPU Architecture
x86_64
CPU Manufacturer
AMD
CPU Family
EPYC
CPU Model
7B13
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
512 KiB
CPU L2 Cache Total
24 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, 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, vmmcall, fsgsbase, tsc_adjust, bmi1, avx2, smep, bmi2, erms, invpcid, rdseed, adx, smap, clflushopt, clwb, sha_ni, xsaveopt, xsavec, xgetbv1, clzero, xsaveerptr, arat, npt, nrip_save, umip, vaes, vpclmulqdq, rdpid, fsrm
Ecpus
44.3
Scalability
46.15

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

      A high-density, memory-optimized virtual server featuring ninety-six dedicated AMD EPYC vCPUs and seven hundred sixty-eight gigabytes of RAM.

      Memory OptimizedGeneral Purpose

      Google Cloud Platform n2d-highmem-96 is a memory-optimized server featuring 96 dedicated vCPUs and 768.0 GB of RAM, powered by an AMD EPYC 7B13 processor on the x86_64 architecture. With 48 physical cores and 2 threads per core, it delivers 8.0 GB of memory per vCPU. The server does not include local storage or GPU accelerators. In benchmark testing, it demonstrates top-tier multi-core CPU performance and encryption capabilities, alongside strong single-core performance, database operations, and Redis throughput. This hardware profile provides qualitative cost efficiency for memory-intensive workloads by offering high resource density without the added cost of bundled storage. It is highly suited for large in-memory databases, heavy web serving, software compilation, and data processing tasks.

      Economics

      Average Price per Region

      Prices per Zone

      Lowest Prices

      Performance

      Workload Profiles

      2.7Score
      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) 6,980,752 ops/sec 1,919,289 ops/sec 3.7 50.00% 50.00% +92.00%
      Redis RPS (pipeline=16, SET) 48,987,218 ops/sec 11,728,058 ops/sec 4.1 20.00% 20.00% +33.00%
      PassMark Memory Mark (composite) 2,848 2,416 1.2 10.00% 10.00% +1.80%
      Memory bandwidth (read, 16 MB ~ L3) 163,645 MB/sec 108,466 MB/sec 1.5 10.00% 10.00% +4.10%
      PassMark single-thread CPU 2,377 Mops/s 2,365 Mops/s 1 10.00% 10.00% 0.00%

      Memory Bandwidth

      Compression

      OpenSSL

      Geekbench Single-Core

      Score: 1,613

      Geekbench Multi-Core

      Score: 16,517

      Passmark CPU Scores

      BENCHMARKSCORE
      Mark
      		 79,745
      Compression
      		 1,511,201
      Encryption
      		 102,726
      Extended Instructions
      		 86,852
      Floating Point Maths
      		 218,941
      Integer Maths
      		 397,525
      Physics
      		 12,680
      Prime Numbers
      		 677
      Single Threaded
      		 2,377
      String Sorting
      		 209,209

      Passmark Memory Scores

      BENCHMARKSCORE
      Memory Mark
      		 2,848
      Database Operations
      		 27,406
      Memory Latency
      		 64
      Memory Read Cached
      		 24,306
      Memory Read Uncached
      		 16,944
      Memory Write
      		 17,790

      Stress-ng Stressors

      CPU integerSingleMultiScaling
      atomic
      		 591.6 158.39 0.27×
      bitops
      		 143,470 7,788,509 54.29×
      branch
      		 164,043 14,416,032 87.88×
      funccall
      		 545.26 33,483 61.41×
      funcret
      		 17,077 1,025,904 60.08×
      hash
      		 139,185 9,906,668 71.18×
      intmath
      		 58,227 3,779,763 64.91×
      nop
      		 284,699 13,678,752 48.05×
      prime
      		 395,101 23,292,406 58.95×
      rotate
      		 77,085 3,705,082 48.06×
      str
      		 801,360 49,972,722 62.36×
      wcs
      		 569,372 30,362,316 53.33×
      CryptographySingleMultiScaling
      crypt
      		 3,334 154,051 46.21×
      CompressionSingleMultiScaling
      jpeg
      		 693.47 46,887 67.61×
      zlib
      		 45.39 4,476 98.60×
      HypervisorSingleMultiScaling
      clock
      		 14,195 172,833 12.18×
      clone
      		 165 5,891 35.70×
      context
      		 1,877 123,985 66.04×
      cpu-sched
      		 58.83 6,118 103.99×
      eventfd
      		 31,013 3,026,309 97.58×
      fault
      		 7,764 156,255 20.13×
      fork
      		 2,541 41,351 16.27×
      getrandom
      		 44,560 2,663,626 59.78×
      hrtimers
      		 375,470 1,709,582 4.55×
      mprotect
      		 439,268 10,941,221 24.91×
      mtx
      		 43,060,183 3,815,982,953 88.62×
      munmap
      		 2,044 52,008 25.45×
      null
      		 198,960 12,738,660 64.03×
      pipe
      		 805,680 19,455,925 24.15×
      priv-instr
      		 298,660 15,011,239 50.26×
      resched
      		 4,577,186 4,330,626 0.95×
      switch
      		 65,287 11,432,975 175.12×
      timermix
      		 138,618 1,746,318 12.60×
      tlb-shootdown
      		 10,615 119,125 11.22×
      tsc
      		 668,381 61,638,753 92.22×
      usersyscall
      		 218,882 12,753,094 58.26×
      vdso
      		 78,677,520 6,204,177,561 78.86×
      vfork
      		 17,849 86,749 4.86×
      x86syscall
      		 2,418,415 144,452,117 59.73×
      yield
      		 39,683,096 38,326,621 0.97×
      zero
      		 1,698,680 107,411,519 63.23×
      CPU float/SIMDSingleMultiScaling
      eigen
      		 14,247 692,933 48.64×
      expmath
      		 2,422 182,044 75.18×
      fma
      		 777,353 57,880,657 74.46×
      fp
      		 227.14 15,772 69.44×
      fractal
      		 6.8 522.09 76.78×
      hyperbolic
      		 3,330 239,430 71.89×
      logmath
      		 3,859 265,932 68.91×
      matrix
      		 2,714 198,870 73.28×
      matrix-3d
      		 800.59 10,265 12.82×
      monte-carlo
      		 117.27 7,209 61.47×
      mpfr
      		 1,401 81,318 58.03×
      powmath
      		 1,193 92,406 77.46×
      trig
      		 1,520 109,949 72.34×
      veccmp
      		 15,979 1,009,431 63.17×
      vecfp
      		 1,476 102,136 69.22×
      vecmath
      		 3,488 264,269 75.76×
      vecshuf
      		 542.63 29,603 54.56×
      vecwide
      		 23,633 1,424,306 60.27×
      AlgorithmsSingleMultiScaling
      bitonicsort
      		 9 441.38 49.04×
      bsearch
      		 264.42 16,743 63.32×
      bubblesort
      		 0.8 56 70.00×
      fibsearch
      		 320.94 18,451 57.49×
      heapsort
      		 9.4 558.19 59.38×
      hsearch
      		 7,919 421,962 53.28×
      insertionsort
      		 11.6 522.79 45.07×
      malloc
      		 2,702,515 90,495,872 33.49×
      mcontend
      		 65.91 12,104 183.64×
      mergesort
      		 27.6 1,931 69.95×
      qsort
      		 17.8 1,253 70.41×
      skiplist
      		 6,177 455,163 73.69×
      tree
      		 1.6 75.29 47.06×
      MemorySingleMultiScaling
      cache
      		 330,126 1,488,847 4.51×
      cachehammer
      		 254,612 10,705,167 42.05×
      cacheline
      		 20.19 2,315 114.67×
      l1cache
      		 157.03 7,183 45.74×
      memcpy
      		 365.85 16,554 45.25×
      memrate
      		 2.33 25.51 10.95×
      memthrash
      		 261.35 1,025 3.92×
      ptr-chase
      		 39,882,873 794,543,501 19.92×
      spinmem
      		 252.18 4,780 18.96×
      stream
      		 47.71 49,544 1038.45×
      vm
      		 57,936 304,101 5.25×

      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
      n2d-highcpu-2 22 GiB0
      n2d-standard-2 28 GiB0
      n2d-highmem-2 216 GiB0
      n2d-highcpu-4 44 GiB0
      n2d-standard-4 416 GiB0
      n2d-highmem-4 432 GiB0
      n2d-highcpu-8 88 GiB0

      Similar Servers

      INSTANCEVENDORvCPUsMEMORYAccelerators

      n2d-highmem-96 FAQs