l40s-360 by OVHcloud

l40s-360 is a Cloud GPU (60 vCPUs, 360 GiB RAM, 400 GB NVMe, 4xL40S 48 GiB VRAM) server offered by OVHcloud with 60 vCPUs, 360 GiB of memory and 400 GB of storage. The pricing starts at 6.2563 USD per hour.
60 vCPU
360 GiB Memory
400 GB Storage
4 GPU
Spare SCore
95,991
(All-cores)
3,048
(Single-core)

Specifications

Server Metadata

Vendor ID
ovh
Name
l40s-360
Description
Cloud GPU (60 vCPUs, 360 GiB RAM, 400 GB NVMe, 4xL40S 48 GiB VRAM)
Family
Cloud GPU
Hw Virt
Status
active
Observed At
2026-11-01T05:48:37.248087

Availability

REGION / IDSPOTONDEMAND
GRA11 (FR) / GRA11
- 6.2563 USD/h

Processor

vCPUs
60
Hypervisor
KVM
CPU Allocation
Dedicated
CPU Cores
60
CPU Speed
2.75 GHz
CPU Architecture
x86_64
CPU Manufacturer
AMD
CPU Family
EPYC
CPU Model
9124
CPU L1D Cache
64 KiB
CPU L1D Cache Total
4 MiB
CPU L1I Cache
64 KiB
CPU L1I Cache Total
4 MiB
CPU L2 Cache
512 KiB
CPU L2 Cache Total
30 MiB
CPU L3 Cache
16 MiB
CPU L3 Cache Total
960 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, 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, svm, cr8_legacy, abm, sse4a, misalignsse, 3dnowprefetch, osvw, perfctr_core, ssbd, ibrs, ibpb, stibp, vmmcall, fsgsbase, tsc_adjust, bmi1, avx2, smep, bmi2, erms, invpcid, avx512f, avx512dq, rdseed, adx, smap, avx512ifma, clflushopt, clwb, avx512cd, sha_ni, avx512bw, avx512vl, xsaveopt, xsavec, xgetbv1, xsaves, avx512_bf16, clzero, xsaveerptr, wbnoinvd, arat, npt, lbrv, nrip_save, tsc_scale, vmcb_clean, pausefilter, pfthreshold, v_vmsave_vmload, vgif, avx512vbmi, umip, pku, ospke, avx512_vbmi2, gfni, vaes, vpclmulqdq, avx512_vnni, avx512_bitalg, avx512_vpopcntdq, la57, rdpid, overflow_recov, succor, fsrm, arch_capabilities
Ecpus
31.5
Scalability
52.5

System Resources and Accelerators

MEMORY
Memory Amount
360 GiB
Memory Amount Actual
352 GiB
ACCELERATORS
Accelerator Count
4
Accelerator Memory Min
48 GiB
Accelerator Memory Total
192 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: 590.48.01, BIOS version: 95.02.66.00.02, Clock rate: 2520)
  • NVIDIA Ada Lovelace NVIDIA L40S (Memory amount: 46068, Firmware version: 590.48.01, BIOS version: 95.02.66.00.02, Clock rate: 2520)
  • NVIDIA Ada Lovelace NVIDIA L40S (Memory amount: 46068, Firmware version: 590.48.01, BIOS version: 95.02.66.00.02, Clock rate: 2520)
  • NVIDIA Ada Lovelace NVIDIA L40S (Memory amount: 46068, Firmware version: 590.48.01, BIOS version: 95.02.66.00.02, Clock rate: 2520)
STORAGE
Storage Size
400 GB
Storage Type
nvme ssd
Storages
  • 400 GB nvme ssd
NETWORK
Network Speed Baseline
25 Gbps
Network Speed Max
25 Gbps
Inbound Traffic
0 GB/month
Outbound Traffic
0 GB/month
IPv4
1

CPU and System Topology

Server Description

A dedicated GPU-accelerated server featuring four NVIDIA L40S GPUs, sixty AMD EPYC cores, and high-speed NVMe storage for parallel computing.

GPU AcceleratedCompute Optimized

OVHcloud l40s-360 is a dedicated GPU-accelerated server featuring 60 AMD EPYC 9124 physical cores running at 2.75 GHz, 360.0 GB of RAM, and 400 GB of local NVMe SSD storage. The hardware configuration is highlighted by four NVIDIA L40S GPUs based on the Ada Lovelace architecture, providing a total of 192 GB of VRAM. Operating on a KVM hypervisor, the server delivers 25 Gbps of baseline network bandwidth. Benchmark data shows strong performance in multi-core CPU execution, ray tracing, and cryptographic operations, placing it in the top 25% of tested systems. Conversely, large-block memory bandwidth and database operations perform in the middle 50%. This resource profile is optimized for parallel computing, software compilation, and GPU-intensive workloads such as rendering and machine learning.

Economics

Average Price per Region

Prices per Zone

Lowest Prices

Performance

Workload Profiles

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): Redis RPS (pipeline=1, SET), Redis RPS (pipeline=16, SET).
Component Score Weight Impact
Raw Reference Normalized Target Actual
Redis RPS (pipeline=1, SET) — 1,918,343 ops/sec — 50.00% 0.00% —
Redis RPS (pipeline=16, SET) — 11,704,734 ops/sec — 20.00% 0.00% —
PassMark Memory Mark (composite) 2,490 2,416 1 10.00% 33.00% 0.00%
Memory bandwidth (read, 16 MB ~ L3) 100,839 MB/sec 108,466 MB/sec 0.91 10.00% 33.00% -3.10%
PassMark single-thread CPU 2,888 Mops/s 2,365 Mops/s 1.2 10.00% 33.00% +6.20%

GPU Bandwidth and Latency

Node (all GPUs)
Node baseline directional bandwidth
66.0 GB/s
Node GPU-to-host bandwidth
76.8 GB/s
Node host-to-GPU bandwidth
97.1 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
723 ns
Benchmark Environment
Driver Version
 535.274.02
CUDA Version (driver)
 12.6
nvbandwidth Version
 v0.5
GPU Count
 4
P2P Supported
No
Efficiency
SM/CE copy efficiency
1.00

Memory Bandwidth

Compression

OpenSSL

Geekbench Single-Core

Score: 1,832

Geekbench Multi-Core

Score: 14,126

Passmark CPU Scores

BENCHMARKSCORE
Mark
 70,151
Compression
 1,173,571
Encryption
 71,133
Extended Instructions
 77,944
Floating Point Maths
 165,861
Integer Maths
 282,926
Physics
 4,810
Prime Numbers
 425
Single Threaded
 2,888
String Sorting
 139,404

Passmark Memory Scores

BENCHMARKSCORE
Memory Mark
 2,490
Database Operations
 21,142
Memory Latency
 90
Memory Read Cached
 27,690
Memory Read Uncached
 23,751
Memory Write
 22,672

Stress-ng Stressors

CPU integerSingleMultiScaling
atomic
 886.29 202.18 0.23×
bitops
 160,553 5,910,374 36.81×
branch
 212,720 11,026,801 51.84×
funccall
 617.8 26,208 42.42×
funcret
 20,920 941,254 44.99×
hash
 162,942 7,350,255 45.11×
intmath
 66,252 2,703,623 40.81×
nop
 617,594 19,439,255 31.48×
prime
 482,572 19,059,082 39.49×
rotate
 88,226 2,756,763 31.25×
str
 1,018,383 40,186,521 39.46×
wcs
 668,839 23,656,962 35.37×
CryptographySingleMultiScaling
crypt
 3,710 124,555 33.58×
CompressionSingleMultiScaling
jpeg
 889.34 23,539 26.47×
zlib
 65.62 3,143 47.89×
HypervisorSingleMultiScaling
clock
 23,849 151,258 6.34×
clone
 164.24 5,941 36.17×
context
 2,878 152,827 53.10×
cpu-sched
 63.18 3,810 60.31×
eventfd
 31,144 1,171,014 37.60×
fault
 10,074 100,389 9.97×
fork
 1,339 39,979 29.87×
getrandom
 52,422 2,135,292 40.73×
hrtimers
 316,145 4,813,820 15.23×
mprotect
 402,951 8,733,065 21.67×
mtx
 85,318,552 3,440,076,348 40.32×
munmap
 1,084 44,621 41.17×
null
 351,274 17,753,487 50.54×
pipe
 210,112 12,347,564 58.77×
priv-instr
 420,921 13,971,065 33.19×
resched
 5,195,047 7,789,319 1.50×
switch
 72,667 1,901,440 26.17×
timermix
 122,557 6,062,194 49.46×
tlb-shootdown
 9,446 94,925 10.05×
tsc
 782,121 44,910,327 57.42×
usersyscall
 316,891 12,454,554 39.30×
vdso
 83,862,008 4,210,866,747 50.21×
vfork
 11,473 79,132 6.90×
x86syscall
 3,984,558 206,338,186 51.78×
yield
 31,469,618 36,926,948 1.17×
zero
 2,756,656 139,620,306 50.65×
CPU float/SIMDSingleMultiScaling
eigen
 16,056 542,675 33.80×
expmath
 2,484 123,640 49.78×
fma
 919,656 43,847,259 47.68×
fp
 251.45 11,533 45.86×
fractal
 7.4 369.63 49.95×
hyperbolic
 3,788 170,820 45.10×
logmath
 4,549 195,678 43.02×
matrix
 4,894 172,175 35.18×
matrix-3d
 435.35 3,158 7.25×
monte-carlo
 171.39 8,468 49.41×
mpfr
 1,697 65,497 38.60×
powmath
 1,317 63,902 48.53×
trig
 1,872 85,374 45.61×
veccmp
 18,964 748,693 39.48×
vecfp
 1,810 80,459 44.44×
vecmath
 4,021 194,767 48.43×
vecshuf
 638.37 24,121 37.78×
vecwide
 28,533 1,069,109 37.47×
AlgorithmsSingleMultiScaling
bitonicsort
 10.4 365.78 35.17×
bsearch
 303.89 12,798 42.12×
bubblesort
 0.8 37 46.25×
fibsearch
 380.85 14,075 36.96×
heapsort
 11 453.78 41.25×
hsearch
 11,290 130,117 11.52×
insertionsort
 13 394.59 30.35×
malloc
 2,641,544 70,092,327 26.53×
mcontend
 49.67 7,587 152.74×
mergesort
 34 1,481 43.57×
qsort
 21.8 907.07 41.61×
skiplist
 7,020 339,601 48.38×
tree
 1.8 38.25 21.25×
MemorySingleMultiScaling
cache
 469,078 844,295 1.80×
cachehammer
 123,091 7,582,379 61.60×
cacheline
 37.58 3,163 84.18×
l1cache
 432.11 14,202 32.87×
memcpy
 448.38 14,269 31.82×
memrate
 3.53 15.24 4.32×
memthrash
 213.08 730.96 3.43×
ptr-chase
 26,394,872 317,765,924 12.04×
spinmem
 223.98 2,672 11.93×
stream
 197.35 45,044 228.24×
vm
 81,132 297,865 3.67×

Stress-ng Raw Scores

Stress-ng Relative Multicore Performance

Static Web Server

Alternatives

Servers of the Same Family

INSTANCEvCPUsMEMORYAccelerators
rtx5000-28 428 GiB1
t1-le-45 845 GiB1
rtx5000-56 856 GiB2
t2-le-45 1545 GiB1
l40s-90 1590 GiB1
a100-180 15180 GiB1
t1-le-90 1690 GiB2

Similar Servers

INSTANCEVENDORvCPUsMEMORYAccelerators

l40s-360 FAQs