logo

h100-380 by OVHcloud

h100-380 is a Cloud GPU (30 vCPUs, 380 GiB RAM, 4040 GB NVMe, 1xH100 80 GiB VRAM) server offered by OVHcloud with 30 vCPUs, 380 GiB of memory and 4.04 TB of storage. The pricing starts at 3.2158 USD per hour.
30 vCPU
380 GiB Memory
4.04 TB Storage
1 GPU
Spare SCore
94563
(All-cores)
3148
(Single-core)

Specifications

Server Metadata

Vendor ID
ovh
Name
h100-380
Description
Cloud GPU (30 vCPUs, 380 GiB RAM, 4040 GB NVMe, 1xH100 80 GiB VRAM)
Family
Cloud GPU
Hw Virt
Status
active
Observed At
2026-08-02T09:19:44.672144

Availability

REGION / IDSPOTONDEMAND
GRA11 (FR) / GRA11
- 3.2158 USD/h
GRA9 (FR) / GRA9
- 3.2158 USD/h

Processor

vCPUs
30
Hypervisor
KVM
CPU Allocation
Dedicated
CPU Cores
30
CPU Speed
3 GHz
CPU Architecture
x86_64
CPU Manufacturer
AMD
CPU Family
EPYC
CPU Model
9354
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
15 MiB
CPU L3 Cache
16 MiB
CPU L3 Cache Total
480 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
30
Scalability
100

System Resources and Accelerators

MEMORY
Memory Amount
380 GiB
Memory Amount Actual
371 GiB
GPU
GPU Count
1
GPU Memory Min
80 GiB
GPU Memory Total
80 GiB
GPU Manufacturer
NVIDIA
GPU Family
Hopper
GPU Model
H100
GPUs
  • NVIDIA Hopper NVIDIA H100 PCIe (Memory amount: 81559, Firmware version: 535.274.02, BIOS version: 96.00.5E.00.01, Clock rate: 1755)
STORAGE
Storage Size
4040 GB
Storage Type
nvme ssd
Storages
  • 200 GB nvme ssd
  • 3840 GB nvme ssd
NETWORK
Network Speed Baseline
8 Gbps
Network Speed Max
8 Gbps
Inbound Traffic
0 GB/month
Outbound Traffic
0 GB/month
IPv4
1

Server Description

A dedicated GPU server combining AMD EPYC processing, local NVMe storage, and NVIDIA H100 acceleration for demanding machine learning workloads.

GPU AcceleratedStorage & DatabaseCompute Optimized

OVHcloud h100-380 is a GPU-accelerated server featuring an AMD EPYC 9354 processor with 30 dedicated physical cores, 380.0 GB of RAM, and a single NVIDIA H100 GPU with 80 GB of VRAM. Built on the x86_64 architecture, this KVM-based instance integrates 4040 GB of local NVMe SSD storage and an 8 Gbps network interface. Performance benchmarks highlight top-tier capabilities in LLM inference and single-core CPU execution, though uncached memory bandwidth for large blocks represents a relative bottleneck. The server delivers high resource density for workloads requiring massive parallel processing and fast local storage. It is engineered for machine learning, large language model inference, data science, and cryptographic operations.

Economics

Average Price per Region

Prices per Zone

Lowest Prices

Performance

Workload Profiles

2.04Score
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) 3,238,250 ops/sec 1,896,415 ops/sec 1.71 50.00% 50.00% +30.80%
Redis RPS (pipeline=16, SET) 29,423,620 ops/sec 11,539,523 ops/sec 2.55 20.00% 20.00% +20.60%
PassMark Memory Mark (composite) 2,891 2,411 1.2 10.00% 10.00% +1.84%
Memory bandwidth (read, 16 MB ~ L3) 923,076 MB/sec 107,368 MB/sec 8.6 10.00% 10.00% +24.00%
PassMark single-thread CPU 2,967 Mops/s 2,363 Mops/s 1.26 10.00% 10.00% +2.34%

Memory Bandwidth

Compression

OpenSSL

Geekbench Single-Core

Score: 2,057

Geekbench Multi-Core

Score: 14,257

Passmark CPU Scores

BENCHMARKSCORE
Mark
 61551
Compression
 809919
Encryption
 40266
Extended Instructions
 75993
Floating Point Maths
 162543
Integer Maths
 185641
Physics
 10793
Prime Numbers
 597
Single Threaded
 2967
String Sorting
 116247

Passmark Memory Scores

BENCHMARKSCORE
Memory Mark
 2891
Database Operations
 21234
Memory Latency
 75
Memory Read Cached
 28489
Memory Read Uncached
 25503
Memory Write
 23480

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

INSTANCEvCPUsMEMORYGPUs
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

INSTANCEVENDORvCPUsMEMORYGPUs

h100-380 FAQs