g5g.2xlarge by Amazon Web Services

g5g.2xlarge is a Graphics intensive [AWS Graviton processors] Gen5 2xlarge server offered by Amazon Web Services with 8 vCPUs, 16 GiB of memory and 0 GB of storage. The pricing starts at 0.0935 USD per hour.
8 vCPU
16 GiB Memory
1 GPU
Spare SCore
12,133
(All-cores)
1,520
(Single-core)

Specifications

Server Metadata

Vendor ID
aws
Name
g5g.2xlarge
Description
Graphics intensive [AWS Graviton processors] Gen5 2xlarge
Family
g5g
Hw Virt
Status
active
Observed At
2026-10-23T15:04:47.222134

Availability

REGION / IDSPOTONDEMAND
Northern Virgina (US) / us-east-1
0.2872 USD/h 0.556 USD/h
Oregon (US) / us-west-2
0.265 USD/h 0.556 USD/h
Aragón (ES) / eu-south-2
0.0935 USD/h 0.6205 USD/h
Seoul (KR) / ap-northeast-2
0.3107 USD/h 0.6839 USD/h
Frankfurt (DE) / eu-central-1
0.2427 USD/h 0.6955 USD/h
Tokyo (JP) / ap-northeast-1
0.3416 USD/h 0.7505 USD/h
Singapore (SG) / ap-southeast-1
0.3705 USD/h 0.778 USD/h

Processor

vCPUs
8
Hypervisor
nitro
CPU Allocation
Dedicated
CPU Cores
8
CPU Speed
2.5 GHz
CPU Architecture
arm64
CPU Manufacturer
AWS
CPU Family
ARMv8
CPU Model
Neoverse-N1
CPU L1D Cache
64 KiB
CPU L1D Cache Total
512 KiB
CPU L1I Cache
64 KiB
CPU L1I Cache Total
512 KiB
CPU L2 Cache
1 MiB
CPU L2 Cache Total
8 MiB
CPU L3 Cache
32 MiB
CPU L3 Cache Total
32 MiB
CPU Flags
fp, asimd, evtstrm, aes, pmull, sha1, sha2, crc32, atomics, fphp, asimdhp, cpuid, asimdrdm, lrcpc, dcpop, asimddp, ssbs
Ecpus
8
Scalability
100

System Resources and Accelerators

MEMORY
Memory Amount
16 GiB
Memory Amount Actual
16 GiB
ACCELERATORS
Accelerator Count
1
Accelerator Memory Min
16 GiB
Accelerator Memory Total
16 GiB
Accelerator Type
GPU
Accelerator Manufacturer
NVIDIA
Accelerator Family
Turing
Accelerator Model
T4G
Accelerators
  • NVIDIA Turing NVIDIA T4G (Memory amount: 15360, Firmware version: 590.48.01, BIOS version: 90.04.AF.00.02, Clock rate: 1590)
STORAGE
Storage Size
0 GB
Storages
    NETWORK
    Network Speed Baseline
    2.5 Gbps
    Network Speed Max
    10 Gbps
    Network Storage Speed Baseline
    2.375 Gbps
    Network Storage Speed Max
    4.75 Gbps
    Inbound Traffic
    0 GB/month
    Outbound Traffic
    0 GB/month
    IPv4
    0

    CPU and System Topology

    Server Description

    An energy-efficient ARM64 virtual machine featuring dedicated NVIDIA GPU acceleration for graphics-intensive workloads and lightweight parallel processing.

    GPU AcceleratedCompute Optimized

    Amazon Web Services g5g.2xlarge is an arm64-based virtual machine featuring 8 dedicated AWS ARMv8 Neoverse-N1 physical cores running at 2.5 GHz, paired with 16.0 GB of system memory. It includes a dedicated NVIDIA Turing T4G GPU with 16 GB of VRAM, though it lacks local storage and provides a baseline network bandwidth of 2 Gbps. Benchmarks show average performance in multi-core processing, software compilation, and ray tracing, but reveal weaknesses in single-core execution, text processing, and cryptographic operations. Due to these performance characteristics, the instance is best suited for graphics-intensive applications, image processing, and lightweight GPU-accelerated workloads. It is less optimal for large language model inference or tasks requiring high cryptographic throughput.

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    0.55Score
    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) 943,066 ops/sec 1,919,289 ops/sec 0.49 50.00% 50.00% -30.00%
    Redis RPS (pipeline=16, SET) 5,808,392 ops/sec 11,728,058 ops/sec 0.48 20.00% 20.00% -14.00%
    PassMark Memory Mark (composite) 1,542 2,416 0.62 10.00% 10.00% -4.70%
    Memory bandwidth (read, 16 MB ~ L3) 137,307 MB/sec 108,466 MB/sec 1.3 10.00% 10.00% +2.70%
    PassMark single-thread CPU 1,110 Mops/s 2,365 Mops/s 0.46 10.00% 10.00% -7.50%

    GPU Bandwidth and Latency

    Node (all GPUs)
    Node baseline directional bandwidth
    		5.7 GB/s
    Node GPU-to-host bandwidth
    		6.6 GB/s
    Node host-to-GPU bandwidth
    		6.6 GB/s
    Single GPU slot
    Per-GPU GPU-to-host bandwidth
    		6.6 GB/s
    Per-GPU host-to-GPU bandwidth
    		6.6 GB/s
    Per-GPU host-device latency
    		780 ns
    Benchmark Environment
    Driver Version
    		 535.183.01
    CUDA Runtime Version
    		 12.6
    nvbandwidth Version
    		 v0.5
    GPU Count
    		 1
    P2P Supported
    No
    Efficiency
    SM/CE copy efficiency
    		1.00

    Memory Bandwidth

    Compression

    OpenSSL

    Geekbench Single-Core

    Score: 950

    Geekbench Multi-Core

    Score: 5,407

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		 5,224
    Compression
    		 37,698
    Encryption
    		 1,082
    Extended Instructions
    		 4,981
    Floating Point Maths
    		 23,970
    Integer Maths
    		 31,568
    Physics
    		 1,113
    Prime Numbers
    		 63
    Single Threaded
    		 1,110
    String Sorting
    		 22,877

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		 1,542
    Database Operations
    		 5,195
    Memory Latency
    		 71
    Memory Read Cached
    		 9,486
    Memory Read Uncached
    		 9,213
    Memory Write
    		 12,547

    Stress-ng Stressors

    CPU integerSingleMultiScaling
    atomic
    		 117.09 113.57 0.97×
    bitops
    		 90,453 721,718 7.98×
    branch
    		 183,300 1,463,464 7.98×
    funccall
    		 1,113 8,890 7.98×
    funcret
    		 17,307 138,052 7.98×
    hash
    		 106,482 850,974 7.99×
    intmath
    		 26,077 209,739 8.04×
    nop
    		 144,767 1,152,831 7.96×
    prime
    		 284,567 2,273,206 7.99×
    rotate
    		 80,350 641,285 7.98×
    str
    		 429,531 3,430,543 7.99×
    wcs
    		 99,256 792,064 7.98×
    CryptographySingleMultiScaling
    crypt
    		 2,423 19,332 7.98×
    CompressionSingleMultiScaling
    jpeg
    		 437.31 3,400 7.77×
    zlib
    		 33.01 215.91 6.54×
    HypervisorSingleMultiScaling
    clock
    		 29,357 186,330 6.35×
    clone
    		 246.63 1,225 4.97×
    context
    		 2,171 17,368 8.00×
    cpu-sched
    		 60.18 552.18 9.18×
    eventfd
    		 55,070 556,695 10.11×
    fault
    		 11,346 77,760 6.85×
    fork
    		 3,818 23,506 6.16×
    getrandom
    		 40,084 320,043 7.98×
    hrtimers
    		 488,256 839,386 1.72×
    mprotect
    		 1,007,573 2,242,655 2.23×
    mtx
    		 9,529,612 311,268,378 32.66×
    munmap
    		 2,642 18,221 6.90×
    null
    		 384,906 3,076,593 7.99×
    pipe
    		 662,514 4,111,386 6.21×
    priv-instr
    		 610,023 4,612,409 7.56×
    resched
    		 2,361,876 939,748 0.40×
    switch
    		 91,291 2,155,902 23.62×
    timermix
    		 275,807 2,209,333 8.01×
    tlb-shootdown
    		 4,555 7,501 1.65×
    vdso
    		 42,696,974 340,400,738 7.97×
    vfork
    		 27,948 118,134 4.23×
    yield
    		 6,177,358 6,418,940 1.04×
    zero
    		 1,722,944 13,790,914 8.00×
    CPU float/SIMDSingleMultiScaling
    eigen
    		 1,111 8,869 7.98×
    expmath
    		 199.66 1,595 7.99×
    fma
    		 532,924 4,257,682 7.99×
    fp
    		 86.2 688.57 7.99×
    fractal
    		 5.4 43 7.96×
    hyperbolic
    		 203.19 1,623 7.99×
    logmath
    		 244.22 1,951 7.99×
    matrix
    		 4,448 35,494 7.98×
    matrix-3d
    		 481.15 3,793 7.88×
    monte-carlo
    		 160.88 1,288 8.00×
    mpfr
    		 533.51 4,261 7.99×
    powmath
    		 146.78 1,172 7.98×
    trig
    		 274.14 2,189 7.98×
    veccmp
    		 14,260 114,083 8.00×
    vecfp
    		 620.52 4,950 7.98×
    vecmath
    		 2,569 20,530 7.99×
    vecshuf
    		 559.29 4,468 7.99×
    vecwide
    		 15,319 122,249 7.98×
    AlgorithmsSingleMultiScaling
    bitonicsort
    		 4.4 35.4 8.05×
    bsearch
    		 214.77 1,715 7.99×
    bubblesort
    		 0.6 4.8 8.00×
    fibsearch
    		 227.56 1,817 7.98×
    heapsort
    		 6.8 54.6 8.03×
    hsearch
    		 4,061 32,495 8.00×
    insertionsort
    		 3.8 30.4 8.00×
    malloc
    		 2,417,889 18,409,974 7.61×
    mcontend
    		 230.91 12,874 55.75×
    mergesort
    		 23.2 185.39 7.99×
    qsort
    		 16.4 129.79 7.91×
    skiplist
    		 4,458 35,616 7.99×
    tree
    		 1 6.4 6.40×
    MemorySingleMultiScaling
    cache
    		 327,171 488,132 1.49×
    cachehammer
    		 74,417 214,147 2.88×
    cacheline
    		 46.84 482.74 10.31×
    l1cache
    		 151.68 1,211 7.98×
    memcpy
    		 369.2 2,946 7.98×
    memrate
    		 3.34 20.34 6.09×
    memthrash
    		 118.92 174.25 1.47×
    ptr-chase
    		 11,447,603 79,450,768 6.94×
    spinmem
    		 400.78 521.36 1.30×
    stream
    		 81.59 3,503 42.93×
    vm
    		 55,202 518,810 9.40×

    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
    g5g.xlarge 48 GiB1
    g5g.4xlarge 1632 GiB1
    g5g.8xlarge 3264 GiB1
    g5g.16xlarge 64128 GiB2
    g5g.metal 64128 GiB2

    Similar Servers

    INSTANCEVENDORvCPUsMEMORYAccelerators

    g5g.2xlarge FAQs