ecs.gn6i-c24g1.24xlarge by Alibaba Cloud

ecs.gn6i-c24g1.24xlarge is a ecs.gn6i family (96 vCPUs, 372 GiB RAM, 0 GB storage, 4xT4 16384 GiB VRAM) server offered by Alibaba Cloud with 96 vCPUs, 372 GiB of memory and 0 GB of storage. The pricing starts at 2.3683 USD per hour.
96 vCPU
372 GiB Memory
4 GPU
Spare SCore
70,574
(All-cores)
1,104
(Single-core)

Specifications

Server Metadata

Vendor ID
alicloud
Name
ecs.gn6i-c24g1.24xlarge
Description
ecs.gn6i family (96 vCPUs, 372 GiB RAM, 0 GB storage, 4xT4 16384 GiB VRAM)
Family
ecs.gn6i
Hw Virt
Average Time To Start
15
Status
active
Observed At
2026-10-24T04:54:00.938512

Availability

REGION / IDSPOTONDEMAND
Virginia (US) / us-east-1
2.9841 USD/h 8.526 USD/h
Tokyo (JP) / ap-northeast-1
2.3683 USD/h 9.473 USD/h
Singapore (SG) / ap-southeast-1
- 9.856 USD/h
Beijing (CN) / cn-beijing
- 10.641 USD/h
Hangzhou (CN) / cn-hangzhou
- 10.641 USD/h
Shenzhen (CN) / cn-shenzhen
- 10.641 USD/h

Processor

vCPUs
96
Hypervisor
KVM
CPU Allocation
Dedicated
CPU Cores
48
CPU Speed
2.5 GHz
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU Family
Xeon
CPU Model
8163
CPU L1D Cache
32 KiB
CPU L1D Cache Total
2 MiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
2 MiB
CPU L2 Cache
1 MiB
CPU L2 Cache Total
48 MiB
CPU L3 Cache
33 MiB
CPU L3 Cache Total
33 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, monitor, ssse3, fma, cx16, pcid, sse4_1, sse4_2, x2apic, movbe, popcnt, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, abm, 3dnowprefetch, pti, fsgsbase, tsc_adjust, bmi1, avx2, smep, bmi2, erms, invpcid, avx512f, avx512dq, rdseed, adx, smap, clflushopt, clwb, avx512cd, avx512bw, avx512vl, xsaveopt, xsavec, xgetbv1, xsaves, arat
Ecpus
63.9
Scalability
66.56

System Resources and Accelerators

MEMORY
Memory Amount
372 GiB
Memory Amount Actual
372 GiB
ACCELERATORS
Accelerator Count
4
Accelerator Memory Min
16 GiB
Accelerator Memory Total
64 GiB
Accelerator Type
GPU
Accelerator Manufacturer
NVIDIA
Accelerator Family
Turing
Accelerator Model
T4
Accelerators
  • NVIDIA Turing Tesla T4 (Memory amount: 16384, Firmware version: 595.71.05, BIOS version: 90.04.38.00.03, Clock rate: 1590)
  • NVIDIA Turing Tesla T4 (Memory amount: 16384, Firmware version: 595.71.05, BIOS version: 90.04.38.00.03, Clock rate: 1590)
  • NVIDIA Turing Tesla T4 (Memory amount: 16384, Firmware version: 595.71.05, BIOS version: 90.04.38.00.03, Clock rate: 1590)
  • NVIDIA Turing Tesla T4 (Memory amount: 16384, Firmware version: 595.71.05, BIOS version: 90.04.38.00.03, Clock rate: 1590)
STORAGE
Storage Size
0 GB
Storages
    NETWORK
    Network Speed Baseline
    30 Gbps
    Inbound Traffic
    0 GB/month
    Outbound Traffic
    0 GB/month
    IPv4
    0

    CPU and System Topology

    Server Description

    Quad NVIDIA T4 GPUs combined with 96 vCPUs and 30 Gbps networking power parallel inference, media processing, and high-throughput compute workloads.

    GPU AcceleratedGeneral Purpose

    Alibaba Cloud ecs.gn6i-c24g1.24xlarge is a GPU-accelerated instance configured with 96 dedicated vCPUs, 48 physical Intel Xeon 8163 cores at 2.5 GHz, 372.0 GB of memory, and four NVIDIA Turing T4 GPUs providing 64 GB total VRAM. Operating on an x86_64 architecture with KVM virtualization, it provides 30 Gbps baseline network bandwidth and relies on remote storage solutions due to the absence of local drives. Benchmark results indicate high multi-core compute throughput, rapid cached memory access, and elevated Redis SET performance, offsetting modest single-core processing metrics. The balance of quad-GPU acceleration, dense vCPU allocation, and high network bandwidth provides a cost-effective hardware foundation for parallelized workloads without needing dedicated local storage drives. This instance is well suited for machine learning inference, parallel data processing, multi-channel media encoding, and high-concurrency memory-cached applications.

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    3.1Score
    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.Partial coverage: missing component benchmark(s): PassMark Memory Mark (composite), PassMark single-thread CPU.
    Component Score Weight Impact
    Raw Reference Normalized Target Actual
    Redis RPS (pipeline=1, SET) 6,766,880 ops/sec 1,919,289 ops/sec 3.6 50.00% 62.00% +120.00%
    Redis RPS (pipeline=16, SET) 39,842,390 ops/sec 11,728,058 ops/sec 3.3 20.00% 25.00% +35.00%
    PassMark Memory Mark (composite) — 2,416 — 10.00% 0.00% —
    Memory bandwidth (read, 16 MB ~ L3) 152,539 MB/sec 108,466 MB/sec 1.4 10.00% 13.00% +4.50%
    PassMark single-thread CPU — 2,365 Mops/s — 10.00% 0.00% —

    Memory Bandwidth

    Compression

    OpenSSL

    Stress-ng Stressors

    CPU integerSingleMultiScaling
    atomic
    		 174.8 94.41 0.54×
    bitops
    		 99,754 5,276,788 52.90×
    branch
    		 121,676 8,860,914 72.82×
    funccall
    		 448.08 22,717 50.70×
    funcret
    		 9,153 636,191 69.50×
    hash
    		 103,319 7,321,169 70.86×
    intmath
    		 17,550 1,293,964 73.73×
    nop
    		 151,169 6,281,544 41.55×
    prime
    		 260,619 13,949,949 53.53×
    rotate
    		 76,387 4,640,032 60.74×
    str
    		 656,854 38,656,802 58.85×
    wcs
    		 422,721 21,873,497 51.74×
    CryptographySingleMultiScaling
    crypt
    		 1,928 90,317 46.84×
    CompressionSingleMultiScaling
    jpeg
    		 471.34 24,980 53.00×
    zlib
    		 41.49 3,208 77.32×
    HypervisorSingleMultiScaling
    clock
    		 13,341 109,244 8.19×
    clone
    		 249.81 6,358 25.45×
    context
    		 1,567 100,138 63.90×
    cpu-sched
    		 72.01 5,694 79.07×
    eventfd
    		 69,010 1,703,470 24.68×
    fault
    		 10,338 60,370 5.84×
    fork
    		 2,236 27,668 12.38×
    getrandom
    		 35,781 2,121,660 59.30×
    hrtimers
    		 133,840 967,953 7.23×
    mprotect
    		 600,422 7,729,493 12.87×
    mtx
    		 10,982,092 1,654,000,559 150.61×
    munmap
    		 1,633 38,930 23.85×
    null
    		 168,135 9,865,134 58.67×
    pipe
    		 800,736 15,168,615 18.94×
    priv-instr
    		 347,966 9,492,676 27.28×
    resched
    		 4,107,076 5,298,643 1.29×
    switch
    		 149,648 5,724,406 38.25×
    timermix
    		 336,252 21,284,329 63.30×
    tlb-shootdown
    		 9,238 87,838 9.51×
    tsc
    		 778,059 52,527,644 67.51×
    usersyscall
    		 242,094 8,498,807 35.11×
    vdso
    		 62,333,448 4,502,771,365 72.24×
    vfork
    		 18,026 41,953 2.33×
    x86syscall
    		 1,923,864 111,556,176 57.99×
    yield
    		 36,659,296 42,150,905 1.15×
    zero
    		 1,267,820 72,148,084 56.91×
    CPU float/SIMDSingleMultiScaling
    eigen
    		 12,892 799,020 61.98×
    expmath
    		 2,115 156,569 74.02×
    fma
    		 477,630 39,695,768 83.11×
    fp
    		 130.94 9,483 72.42×
    fractal
    		 4.2 292.89 69.74×
    hyperbolic
    		 2,198 166,976 75.98×
    logmath
    		 3,606 261,513 72.53×
    matrix
    		 2,370 183,487 77.42×
    matrix-3d
    		 568.06 4,834 8.51×
    monte-carlo
    		 86.19 5,506 63.89×
    mpfr
    		 966.74 50,249 51.98×
    powmath
    		 1,256 92,286 73.48×
    trig
    		 1,353 88,440 65.34×
    veccmp
    		 12,115 577,847 47.70×
    vecfp
    		 1,755 104,254 59.42×
    vecmath
    		 1,657 127,202 76.75×
    vecshuf
    		 737.87 38,535 52.22×
    vecwide
    		 44,638 2,496,309 55.92×
    AlgorithmsSingleMultiScaling
    bitonicsort
    		 5.6 306.79 54.78×
    bsearch
    		 196.8 10,740 54.57×
    bubblesort
    		 0.6 19.2 32.00×
    fibsearch
    		 221.61 13,379 60.37×
    heapsort
    		 6.6 384.19 58.21×
    hsearch
    		 3,953 233,765 59.14×
    insertionsort
    		 6.4 306.8 47.94×
    malloc
    		 1,674,935 50,702,258 30.27×
    mcontend
    		 74.06 18,269 246.67×
    mergesort
    		 21.2 1,282 60.46×
    qsort
    		 15 880.57 58.70×
    skiplist
    		 4,142 279,506 67.48×
    tree
    		 1.2 38.2 31.83×
    MemorySingleMultiScaling
    cache
    		 271,615 904,295 3.33×
    cachehammer
    		 243,927 13,390,394 54.90×
    cacheline
    		 18.15 1,193 65.75×
    l1cache
    		 179.25 7,875 43.93×
    memcpy
    		 301.84 14,716 48.75×
    memrate
    		 0.97 14.93 15.39×
    memthrash
    		 349.83 813.9 2.33×
    ptr-chase
    		 28,857,315 618,651,706 21.44×
    spinmem
    		 284.19 5,392 18.97×
    stream
    		 34.98 33,187 948.75×
    vm
    		 29,703 303,214 10.21×

    Stress-ng Raw Scores

    Stress-ng Relative Multicore Performance

    Static Web Server

    Redis

    Alternatives

    Servers of the Same Family

    INSTANCEvCPUsMEMORYAccelerators
    ecs.gn6i-c4g1.xlarge 415 GiB1
    ecs.gn6i-c8g1.2xlarge 831 GiB1
    ecs.gn6i-c16g1.4xlarge 1662 GiB1
    ecs.gn6i-c24g1.6xlarge 2493 GiB1
    ecs.gn6i-c40g1.10xlarge 40155 GiB1
    ecs.gn6i-c24g1.12xlarge 48186 GiB2

    Similar Servers

    INSTANCEVENDORvCPUsMEMORYAccelerators

    ecs.gn6i-c24g1.24xlarge FAQs