t2-le-90 by OVHcloud

t2-le-90 is a Cloud GPU (30 vCPUs, 90 GiB RAM, 500 GB NVMe, 2xV100S 32 GiB VRAM) server offered by OVHcloud with 30 vCPUs, 90 GiB of memory and 500 GB of storage. The pricing starts at 1.7875 USD per hour.
30 vCPU
90 GiB Memory
500 GB Storage
2 GPU
Spare SCore
43,878
(All-cores)
1,489
(Single-core)

Specifications

Server Metadata

Vendor ID
ovh
Name
t2-le-90
Description
Cloud GPU (30 vCPUs, 90 GiB RAM, 500 GB NVMe, 2xV100S 32 GiB VRAM)
Family
Cloud GPU
Hw Virt
Average Time To Start
261.33
Status
active
Observed At
2026-10-31T15:51:51.805442

Availability

REGION / IDSPOTONDEMAND
BHS5 (CA) / BHS5
- 1.7875 USD/h
DE1 (DE) / DE1
- 1.7875 USD/h
GRA11 (FR) / GRA11
- 1.7875 USD/h
GRA7 (FR) / GRA7
- 1.7875 USD/h
GRA9 (FR) / GRA9
- 1.7875 USD/h
UK1 (GB) / UK1
- 1.7875 USD/h
WAW1 (PL) / WAW1
- 1.7875 USD/h

Processor

vCPUs
30
Hypervisor
KVM
CPU Allocation
Dedicated
CPU Cores
30
CPU Speed
2.9 GHz
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU Family
Xeon
CPU Model
6226R
CPU L1D Cache
32 KiB
CPU L1D Cache Total
960 KiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
960 KiB
CPU L2 Cache
4 MiB
CPU L2 Cache Total
120 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, ss, syscall, nx, pdpe1gb, rdtscp, lm, constant_tsc, arch_perfmon, rep_good, nopl, xtopology, cpuid, tsc_known_freq, pni, pclmulqdq, vmx, ssse3, fma, cx16, pdcm, pcid, sse4_1, sse4_2, x2apic, movbe, popcnt, tsc_deadline_timer, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, abm, 3dnowprefetch, cpuid_fault, ssbd, ibrs, ibpb, stibp, ibrs_enhanced, tpr_shadow, flexpriority, ept, vpid, ept_ad, fsgsbase, tsc_adjust, bmi1, avx2, smep, bmi2, erms, invpcid, mpx, avx512f, avx512dq, rdseed, adx, smap, clflushopt, clwb, avx512cd, avx512bw, avx512vl, xsaveopt, xsavec, xgetbv1, xsaves, arat, vnmi, umip, pku, ospke, avx512_vnni, md_clear, arch_capabilities
Ecpus
29.5
Scalability
98.33

System Resources and Accelerators

MEMORY
Memory Amount
90 GiB
Memory Amount Actual
88 GiB
ACCELERATORS
Accelerator Count
2
Accelerator Memory Min
32 GiB
Accelerator Memory Total
64 GiB
Accelerator Type
GPU
Accelerator Manufacturer
NVIDIA
Accelerator Family
Volta
Accelerator Model
V100S
Accelerators
  • NVIDIA Volta Tesla V100S-PCIE-32GB (Memory amount: 32768, BIOS version: 88.00.98.00.01, Clock rate: 1597)
  • NVIDIA Volta Tesla V100S-PCIE-32GB (Memory amount: 32768, BIOS version: 88.00.98.00.01, Clock rate: 1597)
STORAGE
Storage Size
500 GB
Storage Type
nvme ssd
Storages
  • 500 GB nvme ssd
NETWORK
Network Speed Baseline
4 Gbps
Network Speed Max
4 Gbps
Inbound Traffic
0 GB/month
Outbound Traffic
0 GB/month
IPv4
1

CPU and System Topology

Server Description

A dedicated compute instance featuring dual enterprise GPUs and high-speed NVMe storage for accelerated parallel processing and machine learning inference.

GPU AcceleratedCompute Optimized

OVHcloud t2-le-90 is a GPU-accelerated server instance configured with an Intel Xeon 6226R processor featuring 30 dedicated physical cores running at 2.9 GHz. It is equipped with 90.0 GB of system memory, a 500 GB NVMe SSD, and a 4 Gbps baseline network interface. The primary computational driver is its dual NVIDIA Volta V100S GPU setup, which provides 64 GB of total VRAM. In benchmark evaluations, the instance demonstrates top-tier performance in large language model inference, achieving high token-processing throughput across small, medium, and large models. Standard CPU and memory operations perform in the average range, while multi-threaded decompression tasks show strong results. This hardware profile makes the t2-le-90 highly efficient for parallel computing, machine learning inference, and graphics-intensive workloads, offering a balanced tradeoff between dedicated CPU cores and high-capacity GPU acceleration.

Economics

Average Price per Region

Prices per Zone

Lowest Prices

Performance

Workload Profiles

1.7Score
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) 4,170,330 ops/sec 1,918,343 ops/sec 2.2 50.00% 50.00% +48.00%
Redis RPS (pipeline=16, SET) 23,580,631 ops/sec 11,704,734 ops/sec 2 20.00% 20.00% +15.00%
PassMark Memory Mark (composite) 2,754 2,416 1.2 10.00% 10.00% +1.80%
Memory bandwidth (read, 16 MB ~ L3) 98,513 MB/sec 108,466 MB/sec 0.9 10.00% 10.00% -1.00%
PassMark single-thread CPU 2,108 Mops/s 2,365 Mops/s 0.88 10.00% 10.00% -1.30%

GPU Bandwidth and Latency

Node (all GPUs)
Node baseline directional bandwidth
22.3 GB/s
Node GPU-to-host bandwidth
24.5 GB/s
Node host-to-GPU bandwidth
21.6 GB/s
Single GPU slot
Per-GPU GPU-to-host bandwidth
13.0 GB/s
Per-GPU host-to-GPU bandwidth
11.7 GB/s
Per-GPU host-device latency
737 ns
Benchmark Environment
Driver Version
 535.274.02
CUDA Version (driver)
 12.6
nvbandwidth Version
 v0.5
GPU Count
 2
P2P Supported
No
Efficiency
SM/CE copy efficiency
0.61

Memory Bandwidth

Compression

OpenSSL

Geekbench Single-Core

Score: 1,234

Geekbench Multi-Core

Score: 10,747

Passmark CPU Scores

BENCHMARKSCORE
Mark
 36,761
Compression
 632,481
Encryption
 16,427
Extended Instructions
 45,760
Floating Point Maths
 92,194
Integer Maths
 128,907
Physics
 2,684
Prime Numbers
 214
Single Threaded
 2,108
String Sorting
 72,522

Passmark Memory Scores

BENCHMARKSCORE
Memory Mark
 2,754
Database Operations
 16,812
Memory Latency
 52
Memory Read Cached
 25,536
Memory Read Uncached
 11,295
Memory Write
 10,759

Stress-ng Stressors

CPU integerSingleMultiScaling
atomic
 218.9 174.49 0.80×
bitops
 115,056 3,328,065 28.93×
branch
 163,489 4,691,568 28.70×
funccall
 576.02 16,180 28.09×
funcret
 11,993 345,549 28.81×
hash
 133,991 3,868,236 28.87×
intmath
 23,836 645,395 27.08×
nop
 202,857 5,828,318 28.73×
prime
 332,075 9,403,234 28.32×
rotate
 102,526 2,713,408 26.47×
str
 776,691 21,630,424 27.85×
wcs
 510,423 13,458,185 26.37×
CryptographySingleMultiScaling
crypt
 2,591 74,155 28.62×
CompressionSingleMultiScaling
jpeg
 639.4 14,632 22.88×
zlib
 41.34 1,427 34.53×
HypervisorSingleMultiScaling
clock
 42,738 163,429 3.82×
clone
 382.26 6,459 16.90×
context
 4,707 136,072 28.91×
cpu-sched
 68.24 1,906 27.93×
eventfd
 42,547 1,095,647 25.75×
fault
 12,030 94,120 7.82×
fork
 3,389 50,100 14.78×
getrandom
 49,239 1,407,429 28.58×
hrtimers
 123,638 3,912,802 31.65×
mprotect
 612,281 8,440,426 13.79×
mtx
 17,700,484 1,116,658,654 63.09×
munmap
 2,418 37,313 15.43×
null
 673,321 18,960,958 28.16×
pipe
 781,009 8,295,321 10.62×
priv-instr
 316,586 7,193,990 22.72×
resched
 2,922,975 5,872,961 2.01×
switch
 133,365 3,482,635 26.11×
timermix
 233,423 6,726,457 28.82×
tlb-shootdown
 9,044 69,335 7.67×
tsc
 1,046,758 30,201,303 28.85×
usersyscall
 582,058 8,562,678 14.71×
vdso
 75,722,411 2,176,346,513 28.74×
vfork
 17,653 147,100 8.33×
x86syscall
 7,875,247 226,792,718 28.80×
yield
 31,501,026 41,304,335 1.31×
zero
 3,278,495 93,427,824 28.50×
CPU float/SIMDSingleMultiScaling
eigen
 16,931 485,494 28.68×
expmath
 2,841 82,010 28.86×
fma
 587,961 16,996,354 28.91×
fp
 136.8 3,729 27.26×
fractal
 5.2 146.84 28.24×
hyperbolic
 2,954 85,659 29.00×
logmath
 4,802 137,884 28.71×
matrix
 3,202 92,451 28.87×
matrix-3d
 614.85 3,258 5.30×
monte-carlo
 237.51 6,847 28.83×
mpfr
 1,282 36,068 28.13×
powmath
 1,679 48,580 28.93×
trig
 1,813 51,410 28.36×
veccmp
 16,581 464,565 28.02×
vecfp
 1,386 39,859 28.76×
vecmath
 2,316 66,780 28.83×
vecshuf
 596.05 17,331 29.08×
vecwide
 27,613 657,405 23.81×
AlgorithmsSingleMultiScaling
bitonicsort
 7.6 208.79 27.47×
bsearch
 256.51 7,186 28.01×
bubblesort
 0.8 20.4 25.50×
fibsearch
 257.1 7,423 28.87×
heapsort
 8.8 250.19 28.43×
hsearch
 5,344 143,507 26.85×
insertionsort
 8.58 246.4 28.72×
malloc
 1,697,930 31,367,560 18.47×
mcontend
 70.53 13,337 189.10×
mergesort
 24.8 683.99 27.58×
qsort
 15.8 432.39 27.37×
skiplist
 5,156 146,380 28.39×
tree
 1.4 21.59 15.42×
MemorySingleMultiScaling
cache
 449,354 850,340 1.89×
cachehammer
 303,077 7,395,926 24.40×
cacheline
 15.33 419.11 27.34×
l1cache
 241.19 6,915 28.67×
memcpy
 369.18 10,416 28.21×
memrate
 1.38 10.96 7.94×
memthrash
 167.61 700.76 4.18×
ptr-chase
 27,019,528 251,401,267 9.30×
spinmem
 229.99 1,608 6.99×
stream
 77.61 15,810 203.71×
vm
 56,914 295,862 5.20×

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
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

t2-le-90 FAQs