n4-highmem-4 by Google Cloud Platform

n4-highmem-4 is a 4 vCPUs, 32 GB RAM server offered by Google Cloud Platform with 4 vCPUs, 32 GiB of memory and 0 GB of storage. The pricing starts at 0.03 USD per hour.
4 vCPU
32 GiB Memory
Spare SCore
3,220
(All-cores)
1,565
(Single-core)

Specifications

Server Metadata

Vendor ID
gcp
Server ID
1750010
Name
n4-highmem-4
Description
4 vCPUs, 32 GB RAM
Family
n4
Hw Virt
Average Time To Start
18
Status
active
Observed At
2026-10-31T05:36:32.303782

Availability

REGION / IDSPOTONDEMAND
Columbus (US) / us-east5
0.0759 USD/h 0.238 USD/h
Ashburn (US) / us-east4
0.1272 USD/h 0.238 USD/h
Council Bluffs (US) / us-central1
0.1408 USD/h 0.238 USD/h
The Dalles (US) / us-west1
0.1408 USD/h 0.238 USD/h
Moncks Corner (US) / us-east1
0.1408 USD/h 0.238 USD/h
Mumbai (IN) / asia-south1
0.0512 USD/h 0.2476 USD/h
Delhi (IN) / asia-south2
0.0835 USD/h 0.2476 USD/h
Bangkok (TH) / asia-southeast3
0.0498 USD/h 0.2499 USD/h
Eemshaven (NL) / europe-west4
0.0549 USD/h 0.2499 USD/h
Stockholm (SE) / europe-north2
0.1085 USD/h 0.2499 USD/h
Queretaro (MX) / northamerica-south1
0.0733 USD/h 0.2595 USD/h
Tel Aviv (IL) / me-west1
0.0722 USD/h 0.2618 USD/h
St. Ghislain (BE) / europe-west1
0.0361 USD/h 0.2621 USD/h
Montréal (CA) / northamerica-northeast1
0.0444 USD/h 0.2621 USD/h
Johannesburg (ZA) / africa-south1
0.0629 USD/h 0.2621 USD/h
Toronto (CA) / northamerica-northeast2
0.1017 USD/h 0.2621 USD/h
Hamina (FI) / europe-north1
0.1145 USD/h 0.2621 USD/h
Las Vegas (US) / us-west4
0.1178 USD/h 0.2681 USD/h
London (GB) / europe-west2
0.1095 USD/h 0.2714 USD/h
Changhua County (TW) / asia-east1
0.1048 USD/h 0.2756 USD/h
Milan (IT) / europe-west8
0.0465 USD/h 0.2761 USD/h
Paris (FR) / europe-west9
0.094 USD/h 0.2761 USD/h
Dallas (US) / us-south1
0.052 USD/h 0.2809 USD/h
Madrid (ES) / europe-southwest1
0.0569 USD/h 0.2809 USD/h
Frankfurt (DE) / europe-west3
0.1673 USD/h 0.2809 USD/h
Salt Lake City (US) / us-west3
0.0642 USD/h 0.2861 USD/h
Los Angeles (US) / us-west2
0.1707 USD/h 0.2861 USD/h
Warsaw (PL) / europe-central2
0.1331 USD/h 0.288 USD/h
Doha (QA) / me-central1
0.133 USD/h 0.2892 USD/h
Jurong West (SG) / asia-southeast1
0.1761 USD/h 0.2937 USD/h
Sydney (AU) / australia-southeast1
0.1441 USD/h 0.2975 USD/h
Osaka (JP) / asia-northeast2
0.1166 USD/h 0.3057 USD/h
Seoul (KR) / asia-northeast3
0.1519 USD/h 0.3057 USD/h
Tokyo (JP) / asia-northeast1
0.183 USD/h 0.3057 USD/h
Melbourne (AU) / australia-southeast2
0.1038 USD/h 0.3095 USD/h
Zurich (CH) / europe-west6
0.1509 USD/h 0.3118 USD/h
Jakarta (ID) / asia-southeast2
0.0928 USD/h 0.3201 USD/h
Hong Kong / asia-east2
0.03 USD/h 0.333 USD/h
Santiago (CL) / southamerica-west1
0.0664 USD/h 0.3404 USD/h
Osasco (BR) / southamerica-east1
0.0525 USD/h 0.378 USD/h
Dammam (SA) / me-central2
0.0657 USD/h 0.3809 USD/h

Processor

vCPUs
4
CPU Allocation
Dedicated
CPU Cores
2
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU L1D Cache
48 KiB
CPU L1D Cache Total
96 KiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
64 KiB
CPU L2 Cache
2 MiB
CPU L2 Cache Total
4 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
2.1
Scalability
52.5

System Resources and Accelerators

MEMORY
Memory Amount
32 GiB
Memory Amount Actual
32 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

      Memory-focused virtual instance providing 32 GB RAM across four dedicated vCPUs with advanced vector and matrix extensions for efficient data handling.

      Memory OptimizedGeneral Purpose

      Google Cloud Platform n4-highmem-4 is an x86_64 memory-optimized virtual machine equipped with 4 dedicated vCPUs across 2 physical Intel cores and 32 GB of system RAM. Featuring an 8:1 RAM-to-vCPU ratio and a large 260 MB L3 cache, it supports advanced instruction sets such as AVX-512 and Intel AMX. The instance includes no local storage or discrete GPU accelerators. Benchmarks reveal average single-core CPU execution and static web serving speeds of 7.0 GB/sec, alongside average small-model LLM text generation performance, though multi-core CPU and memory bandwidth metrics remain lower due to the physical core limit. The server provides a cost-effective option for memory-heavy workloads without requiring higher CPU counts, making it suitable for standard in-memory caching, web serving, and lightweight database management.

      Economics

      Average Price per Region

      Prices per Zone

      Lowest Prices

      Performance

      Workload Profiles

      0.41Score
      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) 642,247 ops/sec 1,918,343 ops/sec 0.34 50.00% 50.00% -42.00%
      Redis RPS (pipeline=16, SET) 3,343,129 ops/sec 11,704,734 ops/sec 0.28 20.00% 20.00% -22.00%
      PassMark Memory Mark (composite) 1,689 2,416 0.71 10.00% 10.00% -3.40%
      Memory bandwidth (read, 16 MB ~ L3) 59,818 MB/sec 108,466 MB/sec 0.55 10.00% 10.00% -5.80%
      PassMark single-thread CPU 2,389 Mops/s 2,365 Mops/s 1 10.00% 10.00% 0.00%

      Memory Bandwidth

      Compression

      OpenSSL

      Geekbench Single-Core

      Score: 1,868

      Geekbench Multi-Core

      Score: 4,154

      Passmark CPU Scores

      BENCHMARKSCORE
      Mark
      		 5,956
      Compression
      		 58,109
      Encryption
      		 3,426
      Extended Instructions
      		 4,275
      Floating Point Maths
      		 10,955
      Integer Maths
      		 15,363
      Physics
      		 777
      Prime Numbers
      		 32
      Single Threaded
      		 2,389
      String Sorting
      		 7,880

      Passmark Memory Scores

      BENCHMARKSCORE
      Memory Mark
      		 1,689
      Database Operations
      		 1,874
      Memory Latency
      		 74
      Memory Read Cached
      		 21,450
      Memory Read Uncached
      		 7,061
      Memory Write
      		 6,773

      Stress-ng Stressors

      CPU integerSingleMultiScaling
      atomic
      		 228.97 213.49 0.93×
      bitops
      		 112,098 268,515 2.40×
      branch
      		 106,447 321,593 3.02×
      funccall
      		 553.28 1,450 2.62×
      funcret
      		 9,006 28,369 3.15×
      hash
      		 131,921 376,702 2.86×
      intmath
      		 48,180 117,748 2.44×
      nop
      		 217,470 354,572 1.63×
      prime
      		 383,647 931,478 2.43×
      rotate
      		 79,607 210,412 2.64×
      str
      		 673,356 2,178,586 3.24×
      wcs
      		 538,914 1,405,131 2.61×
      CryptographySingleMultiScaling
      crypt
      		 2,620 6,138 2.34×
      CompressionSingleMultiScaling
      jpeg
      		 611.17 1,644 2.69×
      zlib
      		 35.44 166.51 4.70×
      HypervisorSingleMultiScaling
      clock
      		 41,887 114,528 2.73×
      clone
      		 346.29 630.8 1.82×
      context
      		 3,924 12,984 3.31×
      cpu-sched
      		 66.44 254.84 3.84×
      eventfd
      		 50,388 384,734 7.64×
      fault
      		 18,053 44,227 2.45×
      fork
      		 4,262 14,493 3.40×
      getrandom
      		 42,945 121,769 2.84×
      hrtimers
      		 301,952 482,915 1.60×
      mprotect
      		 916,747 1,128,422 1.23×
      mtx
      		 15,155,426 121,234,992 8.00×
      munmap
      		 3,700 11,746 3.17×
      null
      		 712,056 2,243,117 3.15×
      pipe
      		 608,038 2,839,538 4.67×
      priv-instr
      		 670,332 1,858,024 2.77×
      resched
      		 1,071,926 552,192 0.52×
      switch
      		 94,867 1,051,737 11.09×
      timermix
      		 186,961 631,025 3.38×
      tlb-shootdown
      		 7,457 9,933 1.33×
      tsc
      		 466,649 1,550,988 3.32×
      usersyscall
      		 564,323 1,601,085 2.84×
      vdso
      		 71,970,245 221,836,360 3.08×
      vfork
      		 32,491 94,730 2.92×
      x86syscall
      		 7,961,185 26,796,250 3.37×
      yield
      		 3,609,596 3,926,102 1.09×
      zero
      		 5,164,145 16,989,027 3.29×
      CPU float/SIMDSingleMultiScaling
      eigen
      		 12,557 31,742 2.53×
      expmath
      		 2,587 7,337 2.84×
      fma
      		 939,850 2,408,532 2.56×
      fp
      		 154.77 443.63 2.87×
      fractal
      		 4.8 14 2.92×
      hyperbolic
      		 2,952 8,382 2.84×
      logmath
      		 3,820 12,123 3.17×
      matrix
      		 3,534 8,357 2.36×
      matrix-3d
      		 600.55 1,297 2.16×
      monte-carlo
      		 236.35 763.89 3.23×
      mpfr
      		 1,378 3,283 2.38×
      powmath
      		 1,553 4,740 3.05×
      trig
      		 2,086 5,628 2.70×
      veccmp
      		 19,579 45,996 2.35×
      vecfp
      		 1,221 3,139 2.57×
      vecmath
      		 3,108 8,746 2.81×
      vecshuf
      		 655.42 1,513 2.31×
      vecwide
      		 27,341 67,687 2.48×
      AlgorithmsSingleMultiScaling
      bitonicsort
      		 6.6 16 2.42×
      bsearch
      		 212.41 556.9 2.62×
      bubblesort
      		 0.8 1.6 2.00×
      fibsearch
      		 235.08 582.37 2.48×
      heapsort
      		 7.4 20 2.70×
      hsearch
      		 5,817 14,781 2.54×
      insertionsort
      		 9.4 16.8 1.79×
      malloc
      		 1,764,694 7,183,350 4.07×
      mcontend
      		 141.13 631.48 4.47×
      mergesort
      		 24.2 70.4 2.91×
      qsort
      		 17.2 49 2.85×
      skiplist
      		 4,342 13,771 3.17×
      tree
      		 1.2 4 3.33×
      MemorySingleMultiScaling
      cache
      		 165,847 314,825 1.90×
      cachehammer
      		 164,684 541,407 3.29×
      cacheline
      		 39.35 166.95 4.24×
      l1cache
      		 792.65 644.79 0.81×
      memcpy
      		 352.77 733.5 2.08×
      memrate
      		 0.17 0.62 3.65×
      memthrash
      		 80.1 159.74 1.99×
      ptr-chase
      		 20,481,667 56,339,447 2.75×
      spinmem
      		 205.59 136.59 0.66×
      stream
      		 5.1 48.31 9.47×
      vm
      		 72,183 257,921 3.57×

      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
      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-highcpu-8 816 GiB0
      n4-standard-8 832 GiB0

      Similar Servers

      INSTANCEVENDORvCPUsMEMORYAccelerators

      n4-highmem-4 FAQs