n4-highcpu-48 by Google Cloud Platform

n4-highcpu-48 is a 48 vCPUs, 96 GB RAM server offered by Google Cloud Platform with 48 vCPUs, 96 GiB of memory and 0 GB of storage. The pricing starts at 0.2319 USD per hour.
48 vCPU
96 GiB Memory
Spare SCore
40,404
(All-cores)
1,681
(Single-core)

Specifications

Server Metadata

Vendor ID
gcp
Server ID
1750022
Name
n4-highcpu-48
Description
48 vCPUs, 96 GB RAM
Family
n4
Hw Virt
Average Time To Start
19
Status
active
Observed At
2026-10-25T13:42:08.788715

Availability

REGION / IDSPOTONDEMAND
Columbus (US) / us-east5
0.6174 USD/h 1.837 USD/h
Ashburn (US) / us-east4
0.9818 USD/h 1.837 USD/h
Council Bluffs (US) / us-central1
1.0348 USD/h 1.837 USD/h
The Dalles (US) / us-west1
1.0348 USD/h 1.837 USD/h
Moncks Corner (US) / us-east1
1.0348 USD/h 1.837 USD/h
Mumbai (IN) / asia-south1
0.4159 USD/h 1.9104 USD/h
Delhi (IN) / asia-south2
0.5855 USD/h 1.9104 USD/h
Bangkok (TH) / asia-southeast3
0.3663 USD/h 1.9288 USD/h
Eemshaven (NL) / europe-west4
0.4464 USD/h 1.9288 USD/h
Stockholm (SE) / europe-north2
0.7609 USD/h 1.9288 USD/h
Queretaro (MX) / northamerica-south1
0.5954 USD/h 2.0023 USD/h
Tel Aviv (IL) / me-west1
0.587 USD/h 2.0207 USD/h
St. Ghislain (BE) / europe-west1
0.2651 USD/h 2.0227 USD/h
Montréal (CA) / northamerica-northeast1
0.3614 USD/h 2.0227 USD/h
Johannesburg (ZA) / africa-south1
0.4853 USD/h 2.0227 USD/h
Toronto (CA) / northamerica-northeast2
0.7471 USD/h 2.0227 USD/h
Hamina (FI) / europe-north1
0.8831 USD/h 2.0227 USD/h
Las Vegas (US) / us-west4
0.9566 USD/h 2.069 USD/h
London (GB) / europe-west2
0.8046 USD/h 2.0941 USD/h
Changhua County (TW) / asia-east1
0.8086 USD/h 2.127 USD/h
Milan (IT) / europe-west8
0.3772 USD/h 2.1309 USD/h
Paris (FR) / europe-west9
0.764 USD/h 2.1309 USD/h
Madrid (ES) / europe-southwest1
0.3992 USD/h 2.1676 USD/h
Dallas (US) / us-south1
0.4223 USD/h 2.1676 USD/h
Frankfurt (DE) / europe-west3
1.291 USD/h 2.1676 USD/h
Salt Lake City (US) / us-west3
0.4709 USD/h 2.2082 USD/h
Los Angeles (US) / us-west2
1.2544 USD/h 2.2082 USD/h
Warsaw (PL) / europe-central2
0.9334 USD/h 2.2227 USD/h
Doha (QA) / me-central1
1.0266 USD/h 2.2319 USD/h
Jurong West (SG) / asia-southeast1
1.3589 USD/h 2.2663 USD/h
Sydney (AU) / australia-southeast1
1.112 USD/h 2.2962 USD/h
Osaka (JP) / asia-northeast2
0.8564 USD/h 2.359 USD/h
Seoul (KR) / asia-northeast3
1.2348 USD/h 2.359 USD/h
Tokyo (JP) / asia-northeast1
1.3446 USD/h 2.359 USD/h
Melbourne (AU) / australia-southeast2
0.7628 USD/h 2.388 USD/h
Zurich (CH) / europe-west6
1.226 USD/h 2.4064 USD/h
Jakarta (ID) / asia-southeast2
0.7162 USD/h 2.4702 USD/h
Hong Kong / asia-east2
0.2319 USD/h 2.5697 USD/h
Santiago (CL) / southamerica-west1
0.5394 USD/h 2.6269 USD/h
Osasco (BR) / southamerica-east1
0.4264 USD/h 2.9169 USD/h
Dammam (SA) / me-central2
0.534 USD/h 2.9391 USD/h

Processor

vCPUs
48
CPU Allocation
Dedicated
CPU Cores
24
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU L1D Cache
48 KiB
CPU L1D Cache Total
1 MiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
768 KiB
CPU L2 Cache
2 MiB
CPU L2 Cache Total
48 MiB
CPU L3 Cache
260 MiB
CPU L3 Cache Total
260 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, smx, est, ssse3, sdbg, fma, cx16, xtpr, pdcm, pcid, dca, sse4_1, sse4_2, x2apic, movbe, popcnt, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, abm, 3dnowprefetch, cat_l3, cat_l2, cdp_l3, cdp_l2, ssbd, mba, 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, cqm_llc, cqm_occup_llc, cqm_mbm_total, cqm_mbm_local, avx_vnni, avx512_bf16, wbnoinvd, arat, avx512vbmi, umip, avx512_vbmi2, gfni, vaes, vpclmulqdq, avx512_vnni, avx512_bitalg, avx512_vpopcntdq, la57, rdpid, cldemote, movdiri, movdir64b, fsrm, md_clear, serialize, tsxldtrk, amx_bf16, avx512_fp16, amx_tile, amx_int8, arch_capabilities
Ecpus
24
Scalability
50

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-density, compute-optimized virtual machine featuring advanced Intel matrix extensions for demanding multi-threaded workloads and cryptographic processing.

      Compute OptimizedGeneral Purpose

      Google Cloud Platform n4-highcpu-48 is an x86_64 compute-optimized server featuring 48 vCPUs across 24 physical Intel cores and 96 GB of RAM. Lacking local storage and GPU accelerators, it relies on dedicated CPU allocation with a 260 MB L3 cache and advanced instruction sets like Intel AMX and AVX-512. Benchmarks show strong multi-core capabilities, highlighted by a Geekbench multi-core score of 17,375.0 and a Passmark CPU score of 58,467.72. It excels in uncached large-block memory bandwidth, database operations, and static web serving. However, its 2.0 GB of RAM per core presents a tradeoff for memory-bound tasks, and large-scale LLM text generation is weak without GPU acceleration. This server is highly suited for compute-intensive workloads, cryptography, physics simulations, and web serving.

      Economics

      Average Price per Region

      Prices per Zone

      Lowest Prices

      Performance

      Workload Profiles

      2.8Score
      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) 7,089,160 ops/sec 1,919,289 ops/sec 3.7 50.00% 50.00% +92.00%
      Redis RPS (pipeline=16, SET) 36,788,795 ops/sec 11,728,058 ops/sec 3.1 20.00% 20.00% +25.00%
      PassMark Memory Mark (composite) 2,567 2,416 1.1 10.00% 10.00% +0.96%
      Memory bandwidth (read, 16 MB ~ L3) 363,271 MB/sec 108,466 MB/sec 3.3 10.00% 10.00% +13.00%
      PassMark single-thread CPU 2,708 Mops/s 2,365 Mops/s 1.1 10.00% 10.00% +0.96%

      Memory Bandwidth

      Compression

      OpenSSL

      Geekbench Single-Core

      Score: 1,862

      Geekbench Multi-Core

      Score: 17,375

      Passmark CPU Scores

      BENCHMARKSCORE
      Mark
      		 58,468
      Compression
      		 765,897
      Encryption
      		 46,067
      Extended Instructions
      		 59,128
      Floating Point Maths
      		 148,530
      Integer Maths
      		 209,425
      Physics
      		 10,017
      Prime Numbers
      		 435
      Single Threaded
      		 2,708
      String Sorting
      		 105,594

      Passmark Memory Scores

      BENCHMARKSCORE
      Memory Mark
      		 2,567
      Database Operations
      		 26,021
      Memory Latency
      		 57
      Memory Read Cached
      		 23,952
      Memory Read Uncached
      		 10,170
      Memory Write
      		 9,689

      Stress-ng Stressors

      CPU integerSingleMultiScaling
      atomic
      		 229.87 158.87 0.69×
      bitops
      		 117,787 3,296,522 27.99×
      branch
      		 110,004 3,962,456 36.02×
      funccall
      		 586.13 16,997 29.00×
      funcret
      		 9,894 343,564 34.72×
      hash
      		 136,442 4,550,411 33.35×
      intmath
      		 51,348 1,506,888 29.35×
      nop
      		 216,214 4,156,206 19.22×
      prime
      		 379,034 10,236,549 27.01×
      rotate
      		 79,035 2,382,530 30.15×
      str
      		 737,126 23,784,003 32.27×
      wcs
      		 555,616 14,767,506 26.58×
      CryptographySingleMultiScaling
      crypt
      		 2,677 70,527 26.34×
      CompressionSingleMultiScaling
      jpeg
      		 630.74 19,431 30.81×
      zlib
      		 34.84 1,659 47.63×
      HypervisorSingleMultiScaling
      clock
      		 33,773 339,546 10.05×
      clone
      		 403.55 7,500 18.59×
      context
      		 2,625 126,163 48.05×
      cpu-sched
      		 63.88 3,016 47.21×
      eventfd
      		 50,621 1,120,341 22.13×
      fault
      		 19,555 197,556 10.10×
      fork
      		 3,214 44,760 13.93×
      getrandom
      		 46,868 1,506,070 32.13×
      hrtimers
      		 218,074 5,593,853 25.65×
      mprotect
      		 892,037 11,209,613 12.57×
      mtx
      		 13,201,772 1,347,745,179 102.09×
      munmap
      		 2,342 67,313 28.74×
      null
      		 709,892 25,523,243 35.95×
      pipe
      		 615,939 19,938,265 32.37×
      priv-instr
      		 665,524 14,574,966 21.90×
      resched
      		 13,260,703 3,941,347 0.30×
      switch
      		 73,220 11,123,402 151.92×
      timermix
      		 179,081 7,161,509 39.99×
      tlb-shootdown
      		 10,342 70,527 6.82×
      tsc
      		 463,372 17,898,694 38.63×
      usersyscall
      		 561,448 13,839,995 24.65×
      vdso
      		 71,412,109 2,584,522,408 36.19×
      vfork
      		 20,439 127,429 6.23×
      x86syscall
      		 7,775,843 302,521,945 38.91×
      yield
      		 35,402,011 48,490,841 1.37×
      zero
      		 5,103,270 188,078,670 36.85×
      CPU float/SIMDSingleMultiScaling
      eigen
      		 11,974 381,743 31.88×
      expmath
      		 2,782 91,795 33.00×
      fma
      		 1,032,672 29,864,994 28.92×
      fp
      		 165.45 5,284 31.94×
      fractal
      		 5.2 166.26 31.97×
      hyperbolic
      		 3,205 102,807 32.08×
      logmath
      		 4,109 144,486 35.16×
      matrix
      		 3,845 99,125 25.78×
      matrix-3d
      		 789.09 7,516 9.53×
      monte-carlo
      		 235.85 8,731 37.02×
      mpfr
      		 1,372 34,294 25.00×
      powmath
      		 1,557 55,014 35.34×
      trig
      		 2,061 61,741 29.95×
      veccmp
      		 19,186 484,873 25.27×
      vecfp
      		 1,214 35,526 29.26×
      vecmath
      		 3,108 99,888 32.14×
      vecshuf
      		 646.34 17,237 26.67×
      vecwide
      		 27,405 739,327 26.98×
      AlgorithmsSingleMultiScaling
      bitonicsort
      		 7 184.79 26.40×
      bsearch
      		 228.19 6,273 27.49×
      bubblesort
      		 0.8 19.2 24.00×
      fibsearch
      		 254.16 7,437 29.26×
      heapsort
      		 7.8 231.19 29.64×
      hsearch
      		 6,382 170,972 26.79×
      insertionsort
      		 10 204.4 20.44×
      malloc
      		 2,016,401 59,486,063 29.50×
      mcontend
      		 89.48 7,312 81.71×
      mergesort
      		 24.2 773.78 31.97×
      qsort
      		 17.2 534.58 31.08×
      skiplist
      		 4,317 153,093 35.46×
      tree
      		 1.2 28.76 23.97×
      MemorySingleMultiScaling
      cache
      		 176,667 2,647,325 14.98×
      cachehammer
      		 179,235 2,395,525 13.37×
      cacheline
      		 49.77 3,155 63.40×
      l1cache
      		 861.49 8,082 9.38×
      memcpy
      		 349.43 8,272 23.67×
      memrate
      		 0.17 6.41 37.71×
      memthrash
      		 159.45 1,274 7.99×
      ptr-chase
      		 23,585,457 377,814,918 16.02×
      spinmem
      		 201.79 1,890 9.37×
      stream
      		 5.1 3,780 741.21×
      vm
      		 65,131 298,013 4.58×

      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
      n4-highcpu-2 24 GiB0
      n4-standard-2 28 GiB0
      n4-highmem-2 216 GiB0
      n4-highcpu-4 48 GiB0
      n4-standard-4 416 GiB0
      n4-highmem-4 432 GiB0
      n4-highcpu-8 816 GiB0

      Similar Servers

      INSTANCEVENDORvCPUsMEMORYAccelerators

      n4-highcpu-48 FAQs