GPU-32xCPU-384GB-2xL40S by UpCloud

GPU-32xCPU-384GB-2xL40S is a GPU (32 vCPUs, 384 GiB RAM, 2x L40S) server offered by UpCloud with 32 vCPUs, 384 GiB of memory and 0 GB of storage. The pricing starts at 3.1284 USD per hour.
32 vCPU
384 GiB Memory
2 GPU
Spare SCore
131,938
(All-cores)
4,187
(Single-core)

Specifications

Server Metadata

Vendor ID
upcloud
Name
GPU-32xCPU-384GB-2xL40S
Description
GPU (32 vCPUs, 384 GiB RAM, 2x L40S)
Family
GPU
Hw Virt
Average Time To Start
57
Status
active
Observed At
2026-10-29T10:52:43.488611

Availability

REGION / IDSPOTONDEMAND
Helsinki #2 (FI) / fi-hel2
3.1284 USD/h 4.1749 USD/h

Processor

vCPUs
32
Hypervisor
KVM
CPU Allocation
Shared
CPU Cores
32
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
2 MiB
CPU L1I Cache
64 KiB
CPU L1I Cache Total
2 MiB
CPU L2 Cache
512 KiB
CPU L2 Cache Total
16 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
31.5
Scalability
98.44

System Resources and Accelerators

MEMORY
Memory Amount
384 GiB
Memory Amount Actual
384 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
    45056 GB/month
    IPv4
    1

    CPU and System Topology

    Server Description

    A high-density accelerator instance featuring dual Ada Lovelace GPUs and 384 GB RAM for demanding parallel computing workloads.

    GPU AcceleratedMemory OptimizedCompute Optimized

    UpCloud GPU-32xCPU-384GB-2xL40S is a KVM-hypervised GPU server designed for parallel computing and memory-intensive workloads. It features an AMD EPYC 9575F processor with 32 physical cores, 32 vCPUs, and 384.0 GB of RAM. The server integrates two NVIDIA L40S GPUs from the Ada Lovelace architecture, providing 96 GB of total VRAM, though it lacks local storage. Performance benchmarks place this server in the top 10% for multi-core CPU execution, database operations, and small-model LLM inference, with strong memory bandwidth and medium-model LLM processing. This hardware profile is highly efficient for consolidating heavy compute tasks, making it a strong fit for machine learning inference, data science, ray tracing, and database operations.

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    4.7Score
    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) 12,766,216 ops/sec 1,919,289 ops/sec 6.8 50.00% 50.00% +160.00%
    Redis RPS (pipeline=16, SET) 80,117,110 ops/sec 11,728,058 ops/sec 6.7 20.00% 20.00% +46.00%
    PassMark Memory Mark (composite) 2,688 2,416 1.1 10.00% 10.00% +0.96%
    Memory bandwidth (read, 16 MB ~ L3) 450,148 MB/sec 108,466 MB/sec 4.1 10.00% 10.00% +15.00%
    PassMark single-thread CPU 4,230 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
    		26.8 GB/s
    Per-GPU host-device latency
    		585 ns
    Benchmark Environment
    Driver Version
    		 595.84
    CUDA Version (driver)
    		 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: 2,751

    Geekbench Multi-Core

    Score: 25,149

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		 87,135
    Compression
    		 1,094,058
    Encryption
    		 55,819
    Extended Instructions
    		 96,477
    Floating Point Maths
    		 269,148
    Integer Maths
    		 271,901
    Physics
    		 16,459
    Prime Numbers
    		 1,042
    Single Threaded
    		 4,230
    String Sorting
    		 149,939

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		 2,688
    Database Operations
    		 41,161
    Memory Latency
    		 95
    Memory Read Cached
    		 38,061
    Memory Read Uncached
    		 35,227
    Memory Write
    		 35,430

    Stress-ng Stressors

    CPU integerSingleMultiScaling
    atomic
    		 545.13 407.28 0.75×
    bitops
    		 255,559 7,830,005 30.64×
    branch
    		 202,293 6,149,547 30.40×
    funccall
    		 988.35 30,049 30.40×
    funcret
    		 29,310 897,913 30.63×
    hash
    		 236,878 7,423,459 31.34×
    intmath
    		 87,627 2,771,752 31.63×
    nop
    		 466,048 14,803,325 31.76×
    prime
    		 665,336 19,754,516 29.69×
    rotate
    		 156,632 4,832,635 30.85×
    str
    		 1,682,192 47,893,203 28.47×
    wcs
    		 1,150,733 32,898,107 28.59×
    CryptographySingleMultiScaling
    crypt
    		 6,013 174,456 29.01×
    CompressionSingleMultiScaling
    jpeg
    		 1,366 38,973 28.53×
    zlib
    		 91.29 2,585 28.32×
    HypervisorSingleMultiScaling
    clock
    		 82,776 259,134 3.13×
    clone
    		 415.87 12,066 29.01×
    context
    		 9,809 308,084 31.41×
    cpu-sched
    		 63.83 2,065 32.35×
    eventfd
    		 78,373 1,434,136 18.30×
    fault
    		 43,982 178,024 4.05×
    fork
    		 6,984 77,686 11.12×
    getrandom
    		 110,076 3,292,350 29.91×
    hrtimers
    		 424,590 7,061,475 16.63×
    mprotect
    		 725,249 14,363,651 19.81×
    mtx
    		 57,790,181 2,954,631,265 51.13×
    munmap
    		 5,298 61,231 11.56×
    null
    		 1,618,610 48,556,250 30.00×
    pipe
    		 661,538 20,320,317 30.72×
    priv-instr
    		 1,499,367 17,761,920 11.85×
    resched
    		 14,678,996 14,485,201 0.99×
    switch
    		 101,905 13,991,623 137.30×
    timermix
    		 368,808 10,710,503 29.04×
    tlb-shootdown
    		 21,837 245,735 11.25×
    tsc
    		 975,825 30,855,536 31.62×
    usersyscall
    		 1,208,177 18,598,538 15.39×
    vdso
    		 109,514,725 3,433,921,008 31.36×
    vfork
    		 41,217 185,930 4.51×
    x86syscall
    		 21,117,012 640,968,672 30.35×
    yield
    		 53,917,583 76,000,543 1.41×
    zero
    		 8,704,344 264,770,879 30.42×
    CPU float/SIMDSingleMultiScaling
    eigen
    		 19,483 584,409 30.00×
    expmath
    		 4,418 136,552 30.91×
    fma
    		 1,740,399 52,050,325 29.91×
    fp
    		 226.81 7,198 31.73×
    fractal
    		 9.8 301.63 30.78×
    hyperbolic
    		 6,017 184,796 30.71×
    logmath
    		 6,186 192,138 31.06×
    matrix
    		 10,257 285,138 27.80×
    matrix-3d
    		 412.39 13,684 33.18×
    monte-carlo
    		 506.13 14,940 29.52×
    mpfr
    		 2,660 77,933 29.30×
    powmath
    		 2,037 63,590 31.22×
    trig
    		 2,703 80,677 29.84×
    veccmp
    		 32,427 980,384 30.23×
    vecfp
    		 2,633 81,618 31.00×
    vecmath
    		 5,236 164,263 31.37×
    vecshuf
    		 1,106 32,956 29.80×
    vecwide
    		 44,618 1,288,163 28.87×
    AlgorithmsSingleMultiScaling
    bitonicsort
    		 14.2 415.39 29.25×
    bsearch
    		 412.29 12,063 29.26×
    bubblesort
    		 1.4 44.2 31.57×
    fibsearch
    		 477.64 14,376 30.10×
    heapsort
    		 15.6 453.99 29.10×
    hsearch
    		 14,171 422,044 29.78×
    insertionsort
    		 16.8 512.2 30.49×
    malloc
    		 5,423,685 74,140,825 13.67×
    mcontend
    		 115.47 5,820 50.40×
    mergesort
    		 42.4 1,263 29.78×
    qsort
    		 29 862.77 29.75×
    skiplist
    		 9,219 275,434 29.88×
    tree
    		 2.8 68.53 24.48×
    MemorySingleMultiScaling
    cache
    		 1,310,329 1,648,686 1.26×
    cachehammer
    		 281,376 7,889,396 28.04×
    cacheline
    		 63.56 3,075 48.38×
    l1cache
    		 750.16 22,018 29.35×
    memcpy
    		 902.13 26,275 29.13×
    memrate
    		 1.13 18.8 16.64×
    memthrash
    		 239.23 1,634 6.83×
    ptr-chase
    		 73,438,136 414,290,125 5.64×
    spinmem
    		 465.98 3,236 6.94×
    stream
    		 478.77 54,687 114.22×
    vm
    		 258,967 296,224 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-192GB-1xL40S 16192 GiB1
    GPU-16xCPU-192GB-2xL40S 16192 GiB2
    GPU-20xCPU-256GB-1xL40S 20256 GiB1
    GPU-20xCPU-256GB-2xL40S 20256 GiB2

    Similar Servers

    INSTANCEVENDORvCPUsMEMORYAccelerators

    GPU-32xCPU-384GB-2xL40S FAQs