a3-edgegpu-8g-nolssd by Google Cloud Platform

a3-edgegpu-8g-nolssd is a Accelerator Optimized: 8 NVIDIA H100 GPU, 208 vCPUs, 1872GB RAM, No LocalSSD server offered by Google Cloud Platform with 208 vCPUs, 1872 GiB of memory and 0 GB of storage. The pricing starts at 10.3625 USD per hour.
208 vCPU
1872 GiB Memory
8 GPU
Spare SCore
166,458
(All-cores)
2,047
(Single-core)

Specifications

Server Metadata

Vendor ID
gcp
Server ID
1720368
Name
a3-edgegpu-8g-nolssd
Description
Accelerator Optimized: 8 NVIDIA H100 GPU, 208 vCPUs, 1872GB RAM, No LocalSSD
Family
a3
Hw Virt
Average Time To Start
28.6
Status
active
Observed At
2026-10-29T17:49:51.215575

Availability

REGION / IDSPOTONDEMAND
Council Bluffs (US) / us-central1
52.3076 USD/h 87.8325 USD/h
The Dalles (US) / us-west1
52.3076 USD/h 87.8325 USD/h
Ashburn (US) / us-east4
52.6991 USD/h 87.8325 USD/h
Mumbai (IN) / asia-south1
44.5072 USD/h 91.3457 USD/h
Toronto (CA) / northamerica-northeast2
23.1023 USD/h 96.7123 USD/h
Columbus (US) / us-east5
43.8649 USD/h 98.9257 USD/h
Las Vegas (US) / us-west4
44.1759 USD/h 98.9257 USD/h
London (GB) / europe-west2
17.6121 USD/h 100.129 USD/h
Paris (FR) / europe-west9
38.4148 USD/h 101.8857 USD/h
Frankfurt (DE) / europe-west3
61.7333 USD/h 103.6423 USD/h
Eemshaven (NL) / europe-west4
67.0626 USD/h 111.8172 USD/h
Seoul (KR) / asia-northeast3
57.4228 USD/h 112.7945 USD/h
Jurong West (SG) / asia-southeast1
55.4958 USD/h 113.555 USD/h
Melbourne (AU) / australia-southeast2
10.3625 USD/h 114.1822 USD/h
Tokyo (JP) / asia-northeast1
20.3835 USD/h 125.7886 USD/h

Processor

vCPUs
208
CPU Allocation
Dedicated
CPU Cores
104
CPU Speed
2.7 GHz
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU Family
Xeon
CPU Model
8481C
CPU L1D Cache
48 KiB
CPU L1D Cache Total
5 MiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
3 MiB
CPU L2 Cache
2 MiB
CPU L2 Cache Total
208 MiB
CPU L3 Cache
105 MiB
CPU L3 Cache Total
210 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, vmx, ssse3, fma, cx16, pcid, sse4_1, sse4_2, x2apic, movbe, popcnt, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, abm, 3dnowprefetch, ssbd, ibrs, ibpb, stibp, ibrs_enhanced, tpr_shadow, flexpriority, ept, vpid, ept_ad, fsgsbase, tsc_adjust, bmi1, avx2, smep, bmi2, erms, invpcid, rtm, avx512f, avx512dq, rdseed, adx, smap, avx512ifma, clflushopt, clwb, avx512cd, sha_ni, avx512bw, avx512vl, xsaveopt, xsavec, xgetbv1, xsaves, avx_vnni, avx512_bf16, arat, vnmi, 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
81.3
Scalability
39.09

System Resources and Accelerators

MEMORY
Memory Amount
2 TB
Memory Amount Actual
2 TB
ACCELERATORS
Accelerator Count
8
Accelerator Memory Min
80 GiB
Accelerator Memory Total
640 GiB
Accelerator Type
GPU
Accelerator Manufacturer
NVIDIA
Accelerator Family
Hopper
Accelerator Model
H100
Accelerators
  • NVIDIA Hopper NVIDIA H100 80GB HBM3 (Memory amount: 81559, Firmware version: 595.71.05, BIOS version: 96.00.CF.00.01, Clock rate: 1980)
  • NVIDIA Hopper NVIDIA H100 80GB HBM3 (Memory amount: 81559, Firmware version: 595.71.05, BIOS version: 96.00.CF.00.01, Clock rate: 1980)
  • NVIDIA Hopper NVIDIA H100 80GB HBM3 (Memory amount: 81559, Firmware version: 595.71.05, BIOS version: 96.00.CF.00.01, Clock rate: 1980)
  • NVIDIA Hopper NVIDIA H100 80GB HBM3 (Memory amount: 81559, Firmware version: 595.71.05, BIOS version: 96.00.CF.00.01, Clock rate: 1980)
  • NVIDIA Hopper NVIDIA H100 80GB HBM3 (Memory amount: 81559, Firmware version: 595.71.05, BIOS version: 96.00.CF.00.01, Clock rate: 1980)
  • NVIDIA Hopper NVIDIA H100 80GB HBM3 (Memory amount: 81559, Firmware version: 595.71.05, BIOS version: 96.00.CF.00.01, Clock rate: 1980)
  • NVIDIA Hopper NVIDIA H100 80GB HBM3 (Memory amount: 81559, Firmware version: 595.71.05, BIOS version: 96.00.CF.00.01, Clock rate: 1980)
  • NVIDIA Hopper NVIDIA H100 80GB HBM3 (Memory amount: 81559, Firmware version: 595.71.05, BIOS version: 96.00.CF.00.01, Clock rate: 1980)
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 accelerator-optimized platform featuring eight NVIDIA H100 GPUs and massive memory bandwidth for high-scale parallel computing and machine learning.

    GPU AcceleratedMemory OptimizedCompute Optimized

    Google Cloud Platform a3-edgegpu-8g-nolssd is an accelerator-optimized server featuring 208 dedicated vCPUs from an Intel Xeon 8481C processor, 1872.0 GB of RAM, and eight NVIDIA H100 Hopper GPUs with 640 GB of total VRAM. It contains no local SSD storage. Benchmarks place its multi-core CPU performance, memory bandwidth, and LLM inference prompt processing in the top 10% tier, while single-core performance remains in the average middle 50% range. The lack of local storage represents a key architectural tradeoff, requiring external storage solutions. This server is designed for heavy parallel computing workloads, including large-scale machine learning, LLM inference, and ray tracing.

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    6.3Score
    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) 22,557,621 ops/sec 1,919,289 ops/sec 12 50.00% 50.00% +250.00%
    Redis RPS (pipeline=16, SET) 113,037,618 ops/sec 11,728,058 ops/sec 9.2 20.00% 20.00% +56.00%
    PassMark Memory Mark (composite) 2,168 2,416 0.92 10.00% 10.00% -0.83%
    Memory bandwidth (read, 16 MB ~ L3) 454,651 MB/sec 108,466 MB/sec 4.1 10.00% 10.00% +15.00%
    PassMark single-thread CPU 3,072 Mops/s 2,365 Mops/s 1.3 10.00% 10.00% +2.70%

    GPU Bandwidth and Latency

    Node (all GPUs)
    Node baseline directional bandwidth
    		96.9 GB/s
    Node GPU-to-host bandwidth
    		114.2 GB/s
    Node host-to-GPU bandwidth
    		111.1 GB/s
    Peer-to-peer
    GPU-to-GPU P2P duplex bandwidth
    		393.0 GB/s
    GPU-to-GPU P2P gather bandwidth
    		399.7 GB/s
    GPU-to-GPU P2P latency
    		827 ns
    GPU-to-GPU P2P bandwidth
    		397.4 GB/s
    Single GPU slot
    Per-GPU GPU-to-host bandwidth
    		28.6 GB/s
    Per-GPU host-to-GPU bandwidth
    		27.7 GB/s
    Per-GPU host-device latency
    		1,172 ns
    Per-GPU device memory copy bandwidth
    		1490.7 GB/s
    Benchmark Environment
    Driver Version
    		 595.71.05
    CUDA Version (driver)
    		 13.2
    nvbandwidth Version
    		 v0.8
    GPU Count
    		 8
    P2P Supported
    Yes
    Efficiency
    SM/CE copy efficiency
    		0.91

    Memory Bandwidth

    Compression

    OpenSSL

    Geekbench Single-Core

    Score: 2,109

    Geekbench Multi-Core

    Score: 18,961

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		 120,168
    Compression
    		 2,482,042
    Encryption
    		 160,627
    Extended Instructions
    		 163,937
    Floating Point Maths
    		 565,765
    Integer Maths
    		 794,187
    Physics
    		 14,269
    Prime Numbers
    		 935
    Single Threaded
    		 3,072
    String Sorting
    		 315,345

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		 2,168
    Database Operations
    		 32,473
    Memory Latency
    		 81
    Memory Read Cached
    		 27,997
    Memory Read Uncached
    		 9,285
    Memory Write
    		 8,207

    Stress-ng Stressors

    CPU integerSingleMultiScaling
    atomic
    		 176.35 35.28 0.20×
    bitops
    		 157,615 14,627,476 92.81×
    branch
    		 146,260 17,764,976 121.46×
    funccall
    		 790.3 72,611 91.88×
    funcret
    		 13,166 1,558,538 118.38×
    hash
    		 188,763 20,879,167 110.61×
    intmath
    		 68,284 6,369,947 93.29×
    nop
    		 280,400 18,121,747 64.63×
    prime
    		 497,972 40,713,053 81.76×
    rotate
    		 105,733 8,937,386 84.53×
    str
    		 993,668 88,710,348 89.28×
    wcs
    		 699,007 56,437,148 80.74×
    CryptographySingleMultiScaling
    crypt
    		 3,363 253,582 75.40×
    CompressionSingleMultiScaling
    jpeg
    		 861.38 73,004 84.75×
    zlib
    		 66.9 7,683 114.85×
    HypervisorSingleMultiScaling
    clock
    		 22,453 198,000 8.82×
    clone
    		 403.61 12,935 32.05×
    context
    		 5,497 716,174 130.29×
    cpu-sched
    		 66.51 12,351 185.69×
    eventfd
    		 68,760 2,425,281 35.27×
    fault
    		 31,661 161,156 5.09×
    fork
    		 3,915 51,952 13.27×
    getrandom
    		 62,301 6,494,276 104.24×
    hrtimers
    		 150,422 15,658,274 104.10×
    mprotect
    		 622,136 14,259,427 22.92×
    mtx
    		 25,060,765 4,688,825,962 187.10×
    munmap
    		 2,986 62,058 20.78×
    null
    		 948,214 110,277,300 116.30×
    pipe
    		 619,539 90,269,348 145.70×
    priv-instr
    		 878,196 7,358,020 8.38×
    resched
    		 9,583,170 24,452,751 2.55×
    switch
    		 93,306 6,610,080 70.84×
    timermix
    		 231,332 28,065,211 121.32×
    tlb-shootdown
    		 7,069 146,443 20.72×
    tsc
    		 611,033 79,579,587 130.24×
    usersyscall
    		 745,968 7,356,814 9.86×
    vdso
    		 90,965,457 10,438,348,086 114.75×
    vfork
    		 22,305 116,738 5.23×
    x86syscall
    		 10,562,050 1,369,068,695 129.62×
    yield
    		 158,087,318 181,645,744 1.15×
    zero
    		 6,814,672 824,078,005 120.93×
    CPU float/SIMDSingleMultiScaling
    eigen
    		 17,169 1,614,326 94.02×
    expmath
    		 3,732 397,037 106.39×
    fma
    		 1,456,987 140,686,289 96.56×
    fp
    		 323.99 36,476 112.58×
    fractal
    		 7.4 787.79 106.46×
    hyperbolic
    		 4,308 454,716 105.56×
    logmath
    		 5,535 655,969 118.51×
    matrix
    		 6,503 512,598 78.82×
    matrix-3d
    		 640.12 15,570 24.32×
    monte-carlo
    		 311.57 35,771 114.81×
    mpfr
    		 1,830 138,778 75.84×
    powmath
    		 2,060 231,433 112.36×
    trig
    		 2,818 251,209 89.15×
    veccmp
    		 27,301 2,046,152 74.95×
    vecfp
    		 2,407 209,232 86.92×
    vecmath
    		 4,171 425,905 102.11×
    vecshuf
    		 4,072 606,025 148.84×
    vecwide
    		 86,212 7,226,698 83.82×
    AlgorithmsSingleMultiScaling
    bitonicsort
    		 9.4 769.57 81.87×
    bsearch
    		 298.49 28,522 95.55×
    bubblesort
    		 1 83.2 83.20×
    fibsearch
    		 331.13 30,573 92.33×
    heapsort
    		 10.6 977.37 92.20×
    hsearch
    		 8,776 720,708 82.12×
    insertionsort
    		 13.4 851.39 63.54×
    malloc
    		 2,772,757 122,325,002 44.12×
    mcontend
    		 51.27 24,677 481.31×
    mergesort
    		 32.6 3,033 93.02×
    qsort
    		 23.2 2,117 91.26×
    skiplist
    		 5,774 672,356 116.44×
    tree
    		 1.6 108.48 67.80×
    MemorySingleMultiScaling
    cache
    		 262,937 8,574,455 32.61×
    cachehammer
    		 184,671 7,981,286 43.22×
    cacheline
    		 35.97 2,478 68.90×
    l1cache
    		 1,154 23,314 20.21×
    memcpy
    		 467.28 32,241 69.00×
    memrate
    		 0.12 9.83 81.92×
    memthrash
    		 480.34 1,134 2.36×
    ptr-chase
    		 26,623,869 1,429,928,649 53.71×
    spinmem
    		 253.19 8,387 33.12×
    stream
    		 6.65 39,898 5999.72×
    vm
    		 128,466 311,077 2.42×

    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
    a3-highgpu-1g 26234 GiB1
    a3-highgpu-2g 52468 GiB2
    a3-highgpu-4g 104936 GiB4
    a3-highgpu-8g 2081872 GiB8
    a3-megagpu-8g 2081872 GiB8
    a3-edgegpu-8g 2081872 GiB8
    a3-ultragpu-8g 2242952 GiB8

    Similar Servers

    INSTANCEVENDORvCPUsMEMORYAccelerators

    a3-edgegpu-8g-nolssd FAQs