GPU-16xCPU-192GB-2xL40S by UpCloud

GPU-16xCPU-192GB-2xL40S is a GPU (16 vCPUs, 192 GiB RAM, 2x L40S) server offered by UpCloud with 16 vCPUs, 192 GiB of memory and 0 GB of storage. The pricing starts at 2.2162 USD per hour.
16 vCPU
192 GiB Memory
2 GPU
Spare SCore
67,272
(All-cores)
4,201
(Single-core)

Specifications

Server Metadata

Vendor ID
upcloud
Name
GPU-16xCPU-192GB-2xL40S
Description
GPU (16 vCPUs, 192 GiB RAM, 2x L40S)
Family
GPU
Hw Virt
Average Time To Start
45
Status
active
Observed At
2026-10-23T19:09:53.577385

Availability

REGION / IDSPOTONDEMAND
Helsinki #2 (FI) / fi-hel2
2.2162 USD/h 2.9532 USD/h

Processor

vCPUs
16
Hypervisor
KVM
CPU Allocation
Shared
CPU Cores
16
CPU Speed
2 GHz
CPU Architecture
x86_64
CPU Manufacturer
AMD
CPU Family
EPYC
CPU Model
9575F
CPU L1D Cache
64 KiB
CPU L1D Cache Total
1 MiB
CPU L1I Cache
64 KiB
CPU L1I Cache Total
1 MiB
CPU L2 Cache
512 KiB
CPU L2 Cache Total
8 MiB
CPU L3 Cache
16 MiB
CPU L3 Cache Total
16 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, ht, syscall, nx, mmxext, fxsr_opt, pdpe1gb, rdtscp, lm, rep_good, nopl, cpuid, extd_apicid, 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, cmp_legacy, cr8_legacy, abm, sse4a, misalignsse, 3dnowprefetch, osvw, perfctr_core, ssbd, ibrs, ibpb, stibp, vmmcall, fsgsbase, tsc_adjust, bmi1, avx2, smep, bmi2, invpcid, avx512f, avx512dq, rdseed, adx, smap, avx512ifma, clflushopt, clwb, avx512cd, sha_ni, avx512bw, avx512vl, xsaveopt, xsavec, xgetbv1, xsaves, avx_vnni, avx512_bf16, clzero, xsaveerptr, wbnoinvd, arat, avx512vbmi, umip, pku, ospke, avx512_vbmi2, gfni, vaes, vpclmulqdq, avx512_vnni, avx512_bitalg, avx512_vpopcntdq, la57, rdpid, movdiri, movdir64b, avx512_vp2intersect, arch_capabilities
Ecpus
16
Scalability
100

System Resources and Accelerators

MEMORY
Memory Amount
192 GiB
Memory Amount Actual
192 GiB
ACCELERATORS
Accelerator Count
2
Accelerator Memory Min
48 GiB
Accelerator Memory Total
96 GiB
Accelerator Type
GPU
Accelerator Manufacturer
NVIDIA
Accelerator Family
Ada Lovelace
Accelerator Model
L40S
Accelerators
  • NVIDIA Ada Lovelace NVIDIA L40S (Memory amount: 46068, Firmware version: 595.84, BIOS version: 95.02.66.00.15, Clock rate: 2520)
  • NVIDIA Ada Lovelace NVIDIA L40S (Memory amount: 46068, Firmware version: 595.84, BIOS version: 95.02.66.00.15, Clock rate: 2520)
STORAGE
Storage Size
0 GB
Storages
    NETWORK
    Inbound Traffic
    0 GB/month
    Outbound Traffic
    28672 GB/month
    IPv4
    1

    CPU and System Topology

    Server Description

    A high-density GPU server featuring dual Ada Lovelace accelerators and robust AMD EPYC processing for demanding parallel computing workloads.

    GPU AcceleratedMemory OptimizedCompute Optimized

    UpCloud GPU-16xCPU-192GB-2xL40S is a GPU-accelerated server designed for high-compute and parallel processing workloads. It features 16 vCPUs powered by an AMD EPYC 9575F processor on the x86_64 architecture, paired with 192.0 GB of system memory. The server is equipped with two NVIDIA L40S GPUs from the Ada Lovelace family, delivering a total of 96 GB of VRAM. It does not include local storage, relying instead on external storage solutions. Benchmarks indicate top-tier single-core and multi-core CPU performance, alongside strong memory bandwidth and excellent small-model LLM inference speeds. This configuration is highly suited for machine learning inference, image processing, and data-intensive compute tasks.

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    3.2Score
    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) 6,487,893 ops/sec 1,919,289 ops/sec 3.4 50.00% 50.00% +84.00%
    Redis RPS (pipeline=16, SET) 42,269,238 ops/sec 11,728,058 ops/sec 3.5 20.00% 20.00% +28.00%
    PassMark Memory Mark (composite) 2,829 2,416 1.2 10.00% 10.00% +1.80%
    Memory bandwidth (read, 16 MB ~ L3) 1,081,471 MB/sec 108,466 MB/sec 10 10.00% 10.00% +26.00%
    PassMark single-thread CPU 4,222 Mops/s 2,365 Mops/s 1.8 10.00% 10.00% +6.10%

    GPU Bandwidth and Latency

    Node (all GPUs)
    Node baseline directional bandwidth
    		45.3 GB/s
    Node GPU-to-host bandwidth
    		52.7 GB/s
    Node host-to-GPU bandwidth
    		53.7 GB/s
    Single GPU slot
    Per-GPU GPU-to-host bandwidth
    		26.4 GB/s
    Per-GPU host-to-GPU bandwidth
    		26.9 GB/s
    Per-GPU host-device latency
    		564 ns
    Benchmark Environment
    Driver Version
    		 595.84
    CUDA Runtime Version
    		 13.2
    nvbandwidth Version
    		 v0.8
    GPU Count
    		 2
    P2P Supported
    No
    Efficiency
    SM/CE copy efficiency
    		1.00

    Memory Bandwidth

    Compression

    OpenSSL

    Geekbench Single-Core

    Score: 3,005

    Geekbench Multi-Core

    Score: 22,850

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		 52,577
    Compression
    		 581,688
    Encryption
    		 28,422
    Extended Instructions
    		 52,931
    Floating Point Maths
    		 136,570
    Integer Maths
    		 136,365
    Physics
    		 8,944
    Prime Numbers
    		 536
    Single Threaded
    		 4,222
    String Sorting
    		 81,688

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		 2,829
    Database Operations
    		 22,467
    Memory Latency
    		 88
    Memory Read Cached
    		 38,204
    Memory Read Uncached
    		 35,240
    Memory Write
    		 35,485

    Stress-ng Stressors

    CPU integerSingleMultiScaling
    atomic
    		 590.48 455.6 0.77×
    bitops
    		 255,031 4,038,499 15.84×
    branch
    		 191,159 3,160,267 16.53×
    funccall
    		 1,006 15,815 15.72×
    funcret
    		 29,894 470,848 15.75×
    hash
    		 241,262 3,821,808 15.84×
    intmath
    		 88,459 1,404,083 15.87×
    nop
    		 478,784 7,481,961 15.63×
    prime
    		 680,741 10,568,305 15.52×
    rotate
    		 156,582 2,503,530 15.99×
    str
    		 1,707,226 25,915,604 15.18×
    wcs
    		 1,191,625 17,857,092 14.99×
    CryptographySingleMultiScaling
    crypt
    		 6,035 93,643 15.52×
    CompressionSingleMultiScaling
    jpeg
    		 1,384 21,394 15.46×
    zlib
    		 93.76 1,236 13.18×
    HypervisorSingleMultiScaling
    clock
    		 85,943 284,709 3.31×
    clone
    		 443.21 5,288 11.93×
    context
    		 9,282 152,079 16.38×
    cpu-sched
    		 59.18 1,020 17.23×
    eventfd
    		 100,591 2,746,767 27.31×
    fault
    		 42,791 181,588 4.24×
    fork
    		 7,107 64,038 9.01×
    getrandom
    		 106,370 1,673,122 15.73×
    hrtimers
    		 777,480 2,195,024 2.82×
    mprotect
    		 759,200 8,224,595 10.83×
    mtx
    		 41,760,753 1,534,701,320 36.75×
    munmap
    		 5,729 53,623 9.36×
    null
    		 1,542,735 24,070,170 15.60×
    pipe
    		 846,179 12,851,583 15.19×
    priv-instr
    		 1,456,200 13,603,885 9.34×
    resched
    		 13,843,890 9,475,560 0.68×
    switch
    		 128,661 7,848,812 61.00×
    timermix
    		 362,150 5,495,595 15.17×
    tlb-shootdown
    		 21,902 143,609 6.56×
    tsc
    		 978,573 15,697,605 16.04×
    usersyscall
    		 1,176,433 10,247,672 8.71×
    vdso
    		 108,737,365 1,726,712,475 15.88×
    vfork
    		 43,285 203,262 4.70×
    x86syscall
    		 20,167,405 327,233,395 16.23×
    yield
    		 25,514,539 28,211,324 1.11×
    zero
    		 8,736,793 140,010,863 16.03×
    CPU float/SIMDSingleMultiScaling
    eigen
    		 19,057 307,352 16.13×
    expmath
    		 4,474 70,516 15.76×
    fma
    		 1,735,397 27,490,338 15.84×
    fp
    		 220.8 3,592 16.27×
    fractal
    		 9.8 155.4 15.86×
    hyperbolic
    		 6,032 95,473 15.83×
    logmath
    		 6,273 99,954 15.93×
    matrix
    		 10,258 158,950 15.49×
    matrix-3d
    		 1,538 16,240 10.56×
    monte-carlo
    		 496.77 7,926 15.95×
    mpfr
    		 2,680 41,558 15.51×
    powmath
    		 2,038 32,440 15.91×
    trig
    		 2,713 42,786 15.77×
    veccmp
    		 33,611 524,724 15.61×
    vecfp
    		 2,730 42,828 15.69×
    vecmath
    		 5,253 83,810 15.96×
    vecshuf
    		 1,116 17,454 15.64×
    vecwide
    		 44,881 690,524 15.39×
    AlgorithmsSingleMultiScaling
    bitonicsort
    		 14 217.59 15.54×
    bsearch
    		 407.46 6,281 15.42×
    bubblesort
    		 1.4 22.8 16.29×
    fibsearch
    		 462.25 7,492 16.21×
    heapsort
    		 16 247.59 15.47×
    hsearch
    		 14,956 225,836 15.10×
    insertionsort
    		 17.6 273.4 15.53×
    malloc
    		 5,566,045 37,337,709 6.71×
    mcontend
    		 126.62 2,812 22.21×
    mergesort
    		 43.8 676.38 15.44×
    qsort
    		 29.6 457.79 15.47×
    skiplist
    		 9,341 146,331 15.67×
    tree
    		 2.8 41.58 14.85×
    MemorySingleMultiScaling
    cache
    		 1,323,061 1,487,726 1.12×
    cachehammer
    		 278,674 4,130,345 14.82×
    cacheline
    		 52.77 1,058 20.05×
    l1cache
    		 761.04 11,844 15.56×
    memcpy
    		 910.51 14,311 15.72×
    memrate
    		 1.14 12.8 11.23×
    memthrash
    		 223.06 1,013 4.54×
    ptr-chase
    		 81,270,326 270,609,242 3.33×
    spinmem
    		 452.38 1,415 3.13×
    stream
    		 473.61 28,937 61.10×
    vm
    		 258,011 293,913 1.14×

    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

    INSTANCEvCPUsMEMORYAccelerators
    GPU-8xCPU-64GB-1xL40S 864 GiB1
    GPU-12xCPU-128GB-1xL40S 12128 GiB1
    GPU-12xCPU-128GB-2xL40S 12128 GiB2
    GPU-16xCPU-80GB-1xRTXPRO6000 1680 GiB1
    GPU-16xCPU-192GB-1xL40S 16192 GiB1
    GPU-16xCPU-192GB-3xL40S 16192 GiB3
    GPU-20xCPU-256GB-1xL40S 20256 GiB1

    Similar Servers

    INSTANCEVENDORvCPUsMEMORYAccelerators

    GPU-16xCPU-192GB-2xL40S FAQs