n4-highmem-48 by Google Cloud Platform

n4-highmem-48 is a 48 vCPUs, 384 GB RAM server offered by Google Cloud Platform with 48 vCPUs, 384 GiB of memory and 0 GB of storage. The pricing starts at 0.444 USD per hour.
48 vCPU
384 GiB Memory
Spare SCore
41,375
(All-cores)
1,705
(Single-core)

Specifications

Server Metadata

Vendor ID
gcp
Server ID
1750014
Name
n4-highmem-48
Description
48 vCPUs, 384 GB RAM
Family
n4
Hw Virt
Average Time To Start
18
Status
active
Observed At
2026-09-10T10:51:33.860459

Availability

REGION / IDSPOTONDEMAND
Columbus (US) / us-east5
0.8739 USD/h 2.8565 USD/h
Ashburn (US) / us-east4
1.616 USD/h 2.8565 USD/h
Council Bluffs (US) / us-central1
1.6211 USD/h 2.8565 USD/h
The Dalles (US) / us-west1
1.6211 USD/h 2.8565 USD/h
Moncks Corner (US) / us-east1
1.6211 USD/h 2.8565 USD/h
Delhi (IN) / asia-south2
0.7219 USD/h 2.9707 USD/h
Mumbai (IN) / asia-south1
0.8794 USD/h 2.9707 USD/h
Eemshaven (NL) / europe-west4
0.6684 USD/h 2.9993 USD/h
Bangkok (TH) / asia-southeast3
0.7005 USD/h 2.9993 USD/h
Stockholm (SE) / europe-north2
1.2453 USD/h 2.9993 USD/h
Queretaro (MX) / northamerica-south1
1.259 USD/h 3.1136 USD/h
Tel Aviv (IL) / me-west1
1.0716 USD/h 3.1421 USD/h
St. Ghislain (BE) / europe-west1
0.4817 USD/h 3.1453 USD/h
Montréal (CA) / northamerica-northeast1
0.6259 USD/h 3.1453 USD/h
Johannesburg (ZA) / africa-south1
0.7546 USD/h 3.1453 USD/h
Toronto (CA) / northamerica-northeast2
1.1174 USD/h 3.1453 USD/h
Hamina (FI) / europe-north1
1.4463 USD/h 3.1453 USD/h
Las Vegas (US) / us-west4
1.5815 USD/h 3.2173 USD/h
London (GB) / europe-west2
1.2616 USD/h 3.2564 USD/h
Changhua County (TW) / asia-east1
1.2573 USD/h 3.3075 USD/h
Milan (IT) / europe-west8
0.7236 USD/h 3.3135 USD/h
Paris (FR) / europe-west9
1.0827 USD/h 3.3135 USD/h
Dallas (US) / us-south1
0.6619 USD/h 3.3706 USD/h
Madrid (ES) / europe-southwest1
0.7288 USD/h 3.3706 USD/h
Frankfurt (DE) / europe-west3
1.8252 USD/h 3.3706 USD/h
Salt Lake City (US) / us-west3
0.6086 USD/h 3.4338 USD/h
Los Angeles (US) / us-west2
1.9599 USD/h 3.4338 USD/h
Warsaw (PL) / europe-central2
1.5276 USD/h 3.4563 USD/h
Doha (QA) / me-central1
1.5206 USD/h 3.4706 USD/h
Jurong West (SG) / asia-southeast1
1.9285 USD/h 3.524 USD/h
Sydney (AU) / australia-southeast1
1.6468 USD/h 3.5706 USD/h
Osaka (JP) / asia-northeast2
1.2808 USD/h 3.6683 USD/h
Seoul (KR) / asia-northeast3
1.7498 USD/h 3.6683 USD/h
Tokyo (JP) / asia-northeast1
1.9059 USD/h 3.6683 USD/h
Melbourne (AU) / australia-southeast2
1.195 USD/h 3.7134 USD/h
Zurich (CH) / europe-west6
1.7378 USD/h 3.742 USD/h
Jakarta (ID) / asia-southeast2
1.1136 USD/h 3.8411 USD/h
Hong Kong / asia-east2
0.444 USD/h 3.9959 USD/h
Santiago (CL) / southamerica-west1
1.141 USD/h 4.0848 USD/h
Osasco (BR) / southamerica-east1
0.7382 USD/h 4.5358 USD/h
Dammam (SA) / me-central2
0.837 USD/h 4.5704 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
24.3
Scalability
101.25

System Resources and Accelerators

MEMORY
Memory Amount
384 GiB
Memory Amount Actual
384 GiB
GPU
GPU Count
0
GPU Memory Min
0 MiB
GPU Memory Total
0 MiB
GPUs
    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-memory x86_64 server featuring forty-eight dedicated vCPUs and three hundred eighty-four gigabytes of RAM for memory-intensive enterprise workloads.

      Memory OptimizedGeneral Purpose

      Google Cloud Platform n4-highmem-48 is an x86_64 memory-optimized server configured with 48 dedicated vCPUs (24 physical cores) and 384 GB of RAM, yielding 8 GB of memory per core. Powered by an Intel processor, it features an L3 cache of 260 MB and supports advanced instruction sets including AVX-512 and Intel AMX. The instance lacks bundled local storage and GPUs. Benchmark results indicate strong multi-core CPU performance, high-speed large-block memory bandwidth, and excellent database operation throughput, placing it in the top 25% of tested systems. It also delivers strong performance in static web serving and medium-model LLM prompt processing. This server is optimized for memory-intensive workloads, including enterprise databases, data analytics, and memory-heavy web applications where high memory capacity and robust multi-threaded processing are required.

      Economics

      Average Price per Region

      Prices per Zone

      Lowest Prices

      Performance

      Workload Profiles

      2.75Score
      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) 7,076,530 ops/sec 1,917,397 ops/sec 3.69 50.00% 50.00% +92.10%
      Redis RPS (pipeline=16, SET) 36,475,239 ops/sec 11,681,410 ops/sec 3.12 20.00% 20.00% +25.60%
      PassMark Memory Mark (composite) 2,419 2,418 1 10.00% 10.00% 0.00%
      Memory bandwidth (read, 16 MB ~ L3) 346,514 MB/sec 108,140 MB/sec 3.2 10.00% 10.00% +12.30%
      PassMark single-thread CPU 2,706 Mops/s 2,365 Mops/s 1.14 10.00% 10.00% +1.32%

      Memory Bandwidth

      Compression

      OpenSSL

      Geekbench Single-Core

      Score: 1,811

      Geekbench Multi-Core

      Score: 16,867

      Passmark CPU Scores

      BENCHMARKSCORE
      Mark
      		 58641
      Compression
      		 771212
      Encryption
      		 46410
      Extended Instructions
      		 60303
      Floating Point Maths
      		 148435
      Integer Maths
      		 209228
      Physics
      		 9942
      Prime Numbers
      		 434
      Single Threaded
      		 2706
      String Sorting
      		 105752

      Passmark Memory Scores

      BENCHMARKSCORE
      Memory Mark
      		 2419
      Database Operations
      		 24009
      Memory Latency
      		 51
      Memory Read Cached
      		 22870
      Memory Read Uncached
      		 7691
      Memory Write
      		 7631

      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

      INSTANCEvCPUsMEMORYGPUs
      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

      INSTANCEVENDORvCPUsMEMORYGPUs

      n4-highmem-48 FAQs