ecs.gn5-c8g1.2xlarge by Alibaba Cloud

ecs.gn5-c8g1.2xlarge is a ecs.gn5 family (8 vCPUs, 60 GiB RAM, 472 GB ssd storage, 1xP100 16384 GiB VRAM) server offered by Alibaba Cloud with 8 vCPUs, 60 GiB of memory and 472 GB of storage. The pricing starts at 0.1867 USD per hour.
8 vCPU
60 GiB Memory
472 GB Storage
1 GPU
Spare SCore
4,765
(All-cores)
954
(Single-core)

Specifications

Server Metadata

Vendor ID
alicloud
Name
ecs.gn5-c8g1.2xlarge
Description
ecs.gn5 family (8 vCPUs, 60 GiB RAM, 472 GB ssd storage, 1xP100 16384 GiB VRAM)
Family
ecs.gn5
Hw Virt
Status
active
Observed At
2026-10-24T04:54:00.934440

Availability

REGION / IDSPOTONDEMAND
Virginia (US) / us-east-1
0.1867 USD/h 1.867 USD/h
Frankfurt (DE) / eu-central-1
- 1.942 USD/h
Silicon Valley (US) / us-west-1
0.6778 USD/h 2.054 USD/h
Hong Kong (HK) / cn-hongkong
- 2.067 USD/h
Jakarta (ID) / ap-southeast-5
- 2.176 USD/h
Tokyo (JP) / ap-northeast-1
0.2193 USD/h 2.193 USD/h
Ulanqab (CN) / cn-wulanchabu
0.242 USD/h 2.42 USD/h
Zhangjiakou (CN) / cn-zhangjiakou
0.726 USD/h 2.42 USD/h
Shenzhen (CN) / cn-shenzhen
0.363 USD/h 2.42 USD/h
Beijing (CN) / cn-beijing
- 2.42 USD/h
Hangzhou (CN) / cn-hangzhou
- 2.42 USD/h
Shanghai (CN) / cn-shanghai
- 2.42 USD/h

Processor

vCPUs
8
Hypervisor
KVM
CPU Allocation
Dedicated
CPU Cores
4
CPU Speed
2.5 GHz
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU Family
Xeon
CPU Model
E5-2682 v4
CPU L1D Cache
32 KiB
CPU L1D Cache Total
128 KiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
128 KiB
CPU L2 Cache
256 KiB
CPU L2 Cache Total
1 MiB
CPU L3 Cache
40 MiB
CPU L3 Cache Total
40 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, ssse3, fma, cx16, pcid, sse4_1, sse4_2, x2apic, movbe, popcnt, tsc_deadline_timer, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, abm, 3dnowprefetch, pti, ibrs, ibpb, stibp, fsgsbase, tsc_adjust, bmi1, hle, avx2, smep, bmi2, erms, invpcid, rtm, rdseed, adx, smap, xsaveopt, arat
Ecpus
5
Scalability
62.5

System Resources and Accelerators

MEMORY
Memory Amount
60 GiB
Memory Amount Actual
60 GiB
ACCELERATORS
Accelerator Count
1
Accelerator Memory Min
16 GiB
Accelerator Memory Total
16 GiB
Accelerator Type
GPU
Accelerator Manufacturer
NVIDIA
Accelerator Family
Pascal
Accelerator Model
P100
Accelerators
  • NVIDIA Pascal Tesla P100-PCIE-16GB (Memory amount: 16384, BIOS version: 86.00.41.00.06, Clock rate: 1328)
STORAGE
Storage Size
472 GB
Storage Type
ssd
Storages
    NETWORK
    Network Speed Baseline
    3 Gbps
    Inbound Traffic
    0 GB/month
    Outbound Traffic
    0 GB/month
    IPv4
    0

    CPU and System Topology

    Server Description

    An GPU-accelerated server featuring an NVIDIA P100 graphics processor, local SSD storage, and 60 GB RAM for specialized deep learning inference tasks.

    GPU AcceleratedGeneral Purpose

    Alibaba Cloud ecs.gn5-c8g1.2xlarge is a GPU-accelerated KVM virtual machine configured with an Intel Xeon E5-2682 v4 CPU offering 8 vCPUs across 4 physical cores at 2.5 GHz, paired with 60.0 GB of RAM. The instance includes a single NVIDIA Pascal P100 GPU with 16 GB of VRAM, 472 GB of bundled local SSD storage, and 3 Gbps baseline network bandwidth. Benchmark evaluations highlight weak single-core and multi-core CPU capabilities alongside weak memory read bandwidth due to its older CPU platform. Conversely, the integrated NVIDIA P100 GPU demonstrates strong prompt processing speeds for small and medium LLM inference tasks, reaching top 10% performance for 135M and 7B model prompt ingestion. The combination of local SSD storage, 60 GB RAM, and GPU acceleration provides a resource footprint suitable for deep learning model inference, CUDA batch processing, and GPU-assisted web infrastructure.

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    0.38Score
    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) 605,514 ops/sec 1,919,289 ops/sec 0.32 50.00% 50.00% -43.00%
    Redis RPS (pipeline=16, SET) 3,634,241 ops/sec 11,728,058 ops/sec 0.3 20.00% 20.00% -21.00%
    PassMark Memory Mark (composite) 1,750 2,416 0.75 10.00% 10.00% -2.80%
    Memory bandwidth (read, 16 MB ~ L3) 37,727 MB/sec 108,466 MB/sec 0.35 10.00% 10.00% -10.00%
    PassMark single-thread CPU 1,874 Mops/s 2,365 Mops/s 0.79 10.00% 10.00% -2.30%

    GPU Bandwidth and Latency

    Node (all GPUs)
    Node baseline directional bandwidth
    		11.0 GB/s
    Node GPU-to-host bandwidth
    		13.1 GB/s
    Node host-to-GPU bandwidth
    		11.1 GB/s
    Single GPU slot
    Per-GPU GPU-to-host bandwidth
    		13.1 GB/s
    Per-GPU host-to-GPU bandwidth
    		11.1 GB/s
    Per-GPU host-device latency
    		792 ns
    Benchmark Environment
    Driver Version
    		 580.126.09
    CUDA Runtime Version
    		 13.0
    nvbandwidth Version
    		 v0.8
    GPU Count
    		 1
    P2P Supported
    No
    Efficiency
    SM/CE copy efficiency
    		0.92

    Memory Bandwidth

    Compression

    OpenSSL

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		 6,724
    Compression
    		 94,413
    Encryption
    		 2,028
    Extended Instructions
    		 4,277
    Floating Point Maths
    		 13,217
    Integer Maths
    		 20,889
    Physics
    		 790
    Prime Numbers
    		 54
    Single Threaded
    		 1,874
    String Sorting
    		 13,282

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		 1,750
    Database Operations
    		 2,259
    Memory Latency
    		 72
    Memory Read Cached
    		 20,953
    Memory Read Uncached
    		 6,730
    Memory Write
    		 6,113

    Stress-ng Stressors

    CPU integerSingleMultiScaling
    atomic
    		 233.53 211.94 0.91×
    bitops
    		 101,609 437,079 4.30×
    branch
    		 132,308 780,628 5.90×
    funccall
    		 457.23 1,803 3.94×
    funcret
    		 9,279 52,159 5.62×
    hash
    		 108,039 641,531 5.94×
    intmath
    		 18,248 109,227 5.99×
    nop
    		 154,737 544,420 3.52×
    prime
    		 266,746 1,144,492 4.29×
    rotate
    		 83,819 434,136 5.18×
    str
    		 596,191 2,609,336 4.38×
    wcs
    		 397,277 1,541,996 3.88×
    CryptographySingleMultiScaling
    crypt
    		 1,989 7,639 3.84×
    CompressionSingleMultiScaling
    jpeg
    		 428.46 2,177 5.08×
    zlib
    		 61.67 296.08 4.80×
    HypervisorSingleMultiScaling
    clock
    		 26,504 131,773 4.97×
    clone
    		 419.91 1,407 3.35×
    context
    		 2,657 17,370 6.54×
    cpu-sched
    		 74 518.51 7.01×
    eventfd
    		 30,674 446,745 14.56×
    fault
    		 8,473 48,669 5.74×
    fork
    		 3,584 20,790 5.80×
    getrandom
    		 38,607 178,645 4.63×
    hrtimers
    		 204,972 785,246 3.83×
    mprotect
    		 916,078 1,517,302 1.66×
    mtx
    		 8,694,993 157,858,921 18.16×
    munmap
    		 2,417 13,339 5.52×
    null
    		 307,388 1,969,509 6.41×
    pipe
    		 833,019 2,566,095 3.08×
    priv-instr
    		 529,023 3,010,153 5.69×
    resched
    		 2,045,572 789,382 0.39×
    switch
    		 116,123 1,017,137 8.76×
    timermix
    		 215,683 1,322,068 6.13×
    tlb-shootdown
    		 9,889 17,482 1.77×
    tsc
    		 872,513 6,436,476 7.38×
    usersyscall
    		 377,780 2,192,367 5.80×
    vdso
    		 75,018,197 421,736,326 5.62×
    vfork
    		 19,114 85,647 4.48×
    x86syscall
    		 3,691,256 23,695,870 6.42×
    yield
    		 4,786,916 4,937,609 1.03×
    zero
    		 1,868,335 10,156,735 5.44×
    CPU float/SIMDSingleMultiScaling
    eigen
    		 12,727 63,338 4.98×
    expmath
    		 2,191 12,535 5.72×
    fma
    		 485,834 3,016,278 6.21×
    fp
    		 110.04 536.48 4.88×
    fractal
    		 4.4 22.4 5.09×
    hyperbolic
    		 2,477 14,851 6.00×
    logmath
    		 3,915 22,476 5.74×
    matrix
    		 3,047 13,821 4.54×
    matrix-3d
    		 667.67 1,288 1.93×
    monte-carlo
    		 140.17 859.46 6.13×
    mpfr
    		 992.72 4,061 4.09×
    powmath
    		 1,336 7,579 5.67×
    trig
    		 1,567 6,861 4.38×
    veccmp
    		 12,024 48,741 4.05×
    vecfp
    		 723.36 3,659 5.06×
    vecmath
    		 1,883 11,147 5.92×
    vecshuf
    		 472.13 1,996 4.23×
    vecwide
    		 19,615 84,353 4.30×
    AlgorithmsSingleMultiScaling
    bitonicsort
    		 6 28.2 4.70×
    bsearch
    		 195.54 944.25 4.83×
    bubblesort
    		 0.6 1.6 2.67×
    fibsearch
    		 222.77 1,139 5.11×
    heapsort
    		 7 32.4 4.63×
    hsearch
    		 3,795 16,964 4.47×
    insertionsort
    		 6.8 27 3.97×
    malloc
    		 1,664,996 14,297,646 8.59×
    mcontend
    		 148.48 1,800 12.12×
    mergesort
    		 22.4 121.4 5.42×
    qsort
    		 15.6 75.6 4.85×
    skiplist
    		 5,026 26,801 5.33×
    tree
    		 1.2 4.8 4.00×
    MemorySingleMultiScaling
    cache
    		 185,751 356,470 1.92×
    cachehammer
    		 435,564 3,074,777 7.06×
    cacheline
    		 37.48 294.54 7.86×
    l1cache
    		 164.95 656.15 3.98×
    memcpy
    		 362.99 1,497 4.12×
    memrate
    		 0.98 4.62 4.71×
    memthrash
    		 108.28 245.55 2.27×
    ptr-chase
    		 25,826,190 72,077,945 2.79×
    spinmem
    		 226.19 326.36 1.44×
    stream
    		 31.06 1,032 33.22×
    vm
    		 51,623 277,332 5.37×

    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
    ecs.gn5-c4g1.xlarge 430 GiB1
    ecs.gn5-c4g1.2xlarge 860 GiB2
    ecs.gn5-c8g1.4xlarge 16120 GiB2
    ecs.gn5-c28g1.7xlarge 28112 GiB1
    ecs.gn5-c8g1.8xlarge 32240 GiB4
    ecs.gn5-c8g1.14xlarge 54480 GiB8
    ecs.gn5-c28g1.14xlarge 56224 GiB2

    Similar Servers

    INSTANCEVENDORvCPUsMEMORYAccelerators

    ecs.gn5-c8g1.2xlarge FAQs