Skip to main content

OVHcloud C3-4

Milan

Benchmark complete
Finished Oct 12, 2026
Geekbench 7 single-core
1,809
Faster than 89% of 2 vCPU plans
Geekbench 7 multi-core
3,402
Faster than 91% of 2 vCPU plans
Disk 4K IOPS
40,091
fio random read and write, 50/50
Monthly price at test time
$39.42
86 multi-core points per dollar
Same price today

Server

  • Provider
    OVHcloud
  • Plan
    C3-4
  • Location
    Milan
  • vCPU
    2
  • RAM
    4 GB
  • Storage
    50 GB NVMe
  • Network
    250 Mbps
  • OS image
    Ubuntu 24.04
  • Trigger
    Manual
  • Queued at
    Oct 12, 2026, 01:06:45 UTC
Visit OVHcloud

Latest runs of this plan

Full history of this plan

CPU

Version Single-core Multi-core
Geekbench 7 1,809 3,402 Geekbench Browser

Scores from different Geekbench versions aren't comparable with each other. Charts only use Geekbench 7.

Geekbench 7 workloads

Workload Single-core Multi-core
File Compression
1,813
7th of 106
3,142
30th of 90
Navigation
2,004
2nd of 106
Single-core only
HTML5 Browser
1,824
12th of 106
Single-core only
PDF Viewer
1,839
15th of 106
Single-core only
Photo Library
1,638
18th of 106
3,275
35th of 90
Clang
1,707
4th of 106
3,399
36th of 90
Text Processing
1,662
14th of 106
3,151
37th of 90
Asset Compression
1,788
17th of 106
3,554
37th of 90
HDR
1,853
14th of 106
3,327
26th of 90
Photo Editor
1,969
20th of 106
3,825
29th of 90
Ray Tracer
1,794
4th of 106
3,602
37th of 90
Structure from Motion
1,812
16th of 106
Single-core only
Game Physics
1,753
10th of 106
Single-core only
Video Encoder
1,843
12th of 106
Single-core only
Audio Encoder
1,780
19th of 106
Single-core only
Video Player
1,900
14th of 106
Single-core only

Rank among the 106 runs with Geekbench 7 workload scores. Bars are relative to the best run. Geekbench doesn't run every workload multi-core, so some rows only have a single-core score.

Memory, TLS and scheduler

Memory bandwidth (STREAM)

Test Result Runs Spread
Copy higher is better 48 GB/s 3 0.506%
Scale higher is better 36.4 GB/s 3 0.908%
Add higher is better 34.6 GB/s 3 0.78%
Triad higher is better 34.9 GB/s 3 0.327%

Average of 3 runs. Spread is the standard deviation as a share of the average.

TLS and crypto (OpenSSL)

Test Result Runs Spread
RSA 4096 signs higher is better 499/s 3 0.175%
RSA 4096 verifies higher is better 33,074/s 3 0.0318%
SHA-256 higher is better 2.02 GB/s 3 0.798%
AES-256-GCM higher is better 9.04 GB/s 3 0.0248%
ChaCha20-Poly1305 higher is better 4.7 GB/s 3 0.238%

Average of 3 runs. Spread is the standard deviation as a share of the average.

Scheduler latency (schbench)

Test Result Runs Spread
Request latency, median lower is better 4 ms 3 0%
Requests per second, 90th percentile higher is better 495/s 3 0.117%
Request latency, 99.9th percentile lower is better 6.86 ms 3 6.1%
Requests per second, average higher is better 495/s 3 0.0983%

Average of 3 runs. Spread is the standard deviation as a share of the average.

Disk

Block size Read Write Total
4K 78.2 MB/s (20,019 IOPS) 78.4 MB/s (20,072 IOPS) 156.6 MB/s (40,091 IOPS)
64K 1.23 GB/s (20,211 IOPS) 1.24 GB/s (20,317 IOPS) 2.47 GB/s (40,528 IOPS)
512K 1.39 GB/s (2,847 IOPS) 1.46 GB/s (2,999 IOPS) 2.86 GB/s (5,846 IOPS)
1M 1.44 GB/s (1,474 IOPS) 1.54 GB/s (1,572 IOPS) 2.98 GB/s (3,046 IOPS)

fio random read and write, 50/50, run by YABS against the boot disk.

Network

IPv4 transfer speed

Loading...
Target Protocol Send Receive
Clouvider, London, UK (10G) IPv4 243 Mbps 243 Mbps
Eranium, Amsterdam, NL (100G) IPv4 243 Mbps 244 Mbps
Leaseweb, Singapore, SG (10G) IPv4 220 Mbps 201 Mbps
Clouvider, Los Angeles, CA, US (10G) IPv4 79.5 Mbps 228 Mbps
Leaseweb, NYC, NY, US (10G) IPv4 233 Mbps 229 Mbps

Public iperf3 servers vary in capacity and availability. Compare a target only with itself over time.

Datastores

PostgreSQL

fsync speed (pg_test_fsync)

Method Operations per second Time per operation
One 8 kB write
open_datasync 9,770 0.102 ms
fdatasync 9,647 0.104 ms
fsync 7,915 0.126 ms
fsync_writethrough Not supported here n/a
open_sync 8,258 0.121 ms
Two 8 kB writes
open_datasync 4,920 0.203 ms
fdatasync 8,438 0.119 ms
fsync 7,082 0.141 ms
fsync_writethrough Not supported here n/a
open_sync 4,140 0.242 ms
open_sync at different write sizes
1 * 16kB open_sync write 7,836 0.128 ms
2 * 8kB open_sync writes 4,141 0.242 ms
4 * 4kB open_sync writes 2,104 0.475 ms
8 * 2kB open_sync writes 1,141 0.876 ms
16 * 1kB open_sync writes 577 1.73 ms
fsync on a non-write descriptor
write, fsync, close 7,038 0.142 ms
write, close, fsync 7,031 0.142 ms
Non-synced 8 kB writes
write 617,869 0.002 ms

pg_test_fsync against the Docker volume Postgres keeps its data on. fdatasync is Linux's default wal_sync_method.

pgbench at scale 10

Workload Clients Transactions/s Median p95 p99 p99.9 Failed
Read/write (TPC-B like) 1 2,866 0.347 ms 0.388 ms 0.473 ms 1.69 ms 0
Read/write (TPC-B like) 8 6,236 1.21 ms 2.04 ms 2.72 ms 4.36 ms 0
Read/write (TPC-B like) 32 5,843 4.48 ms 12.1 ms 20 ms 32.8 ms 0
Read only 1 34,011 0.03 ms 0.035 ms 0.044 ms 0.057 ms 0
Read only 8 69,600 0.099 ms 0.152 ms 0.192 ms 0.323 ms 0
Read only 32 67,417 0.361 ms 0.659 ms 0.908 ms 1.38 ms 0

Scale 10 (1,000,000 accounts), loaded in 0.97 s. 60 s per run, prepared statements. Latency comes from sampled transaction logs, and p99.9 is left out when fewer than 1,000 were sampled.

pgbench at scale 100

Workload Clients Transactions/s Median p95 p99 p99.9 Failed
Read/write (TPC-B like) 1 2,203 0.396 ms 0.612 ms 1.24 ms 5.32 ms 0
Read/write (TPC-B like) 8 5,249 1.35 ms 2.96 ms 5.22 ms 8.38 ms 0
Read/write (TPC-B like) 32 6,373 4.64 ms 8 ms 12.2 ms 23.9 ms 0
Read only 1 27,896 0.035 ms 0.047 ms 0.064 ms 0.42 ms 0
Read only 8 54,915 0.13 ms 0.207 ms 0.251 ms 0.598 ms 0
Read only 32 52,901 0.5 ms 0.79 ms 1.11 ms 1.91 ms 0

Scale 100 (10,000,000 accounts), loaded in 9.9 s. 60 s per run, prepared statements. Latency comes from sampled transaction logs, and p99.9 is left out when fewer than 1,000 were sampled.

MySQL

sysbench OLTP

Workload Threads Transactions/s Queries/s Median p95 p99 Errors
Read/write 1 572 11,448 1.67 ms 2.26 ms 3.49 ms 0
Read/write 8 1,105 22,108 7.04 ms 10.7 ms 13.7 ms 0
Read/write 32 1,199 23,987 26.7 ms 41.1 ms 54.8 ms 0
Read only 1 1,262 20,189 0.782 ms 0.826 ms 1.18 ms 0
Read only 8 2,293 36,691 3.36 ms 4.57 ms 5.09 ms 0
Read only 32 2,118 33,881 15 ms 19 ms 20.7 ms 0

10 tables of 100,000 rows, loaded in 5.2 s. 60 s per run, one connection per thread.

Redis

memtier

Mix Operations/s Hits/s Median p99 p99.9
1:10 set:get 64,954 59,049 0.591 ms 3.06 ms 4.48 ms
1:1 set:get 64,803 32,401 0.575 ms 3.07 ms 4.51 ms

1,000,000 keys of 32 bytes. 64 connections over 2 threads, 60 s per ratio.

Web frameworks

Framework Fortunes req/s Single query req/s JSON req/s Fortunes p99 at 256 connections
Next.js nextjs 994 3,211 4,651 357 ms
Express express-postgresjs 9,013 13,357 50,966 48.4 ms
Fastify fastify-postgres 8,838 13,923 55,883 46.1 ms
NestJS nestjs 6,988 7,579 30,125 71.3 ms
Nuxt nuxt 8,264 13,521 48,193 58.7 ms
FastAPI fastapi 4,717 6,223 30,118 220 ms
Django django-postgresql 1,766 1,980 4,672 188 ms
Rails rails 3,013 4,017 16,813 150 ms
Laravel laravel 1,061 1,122 2,098 250 ms
Symfony symfony 1,286 1,401 3,895 205 ms

Peak requests per second across the 64 and 256 connection levels, measured with wrk. Each test has to pass the TechEmpower response checks first, or it's marked invalid.

WooCommerce

Page mix

Step Requests/s Median p95 p99 Error rate
8 virtual users
Home page 3.35 472 ms 518 ms 573 ms 0%
Shop page 3.35 635 ms 689 ms 718 ms 0%
Category page 3.35 406 ms 634 ms 665 ms 0%
Product page 3.35 439 ms 500 ms 525 ms 0%
Product search 3.35 385 ms 441 ms 467 ms 0%
32 virtual users
Home page 3.32 1.75 s 1.98 s 2.17 s 0%
Shop page 3.32 2.33 s 2.61 s 2.67 s 0%
Category page 3.32 2.02 s 2.37 s 2.43 s 0%
Product page 3.32 1.79 s 1.95 s 2.01 s 0%
Product search 3.32 1.66 s 1.78 s 1.88 s 0%

Cart and checkout

Step Requests/s Median p95 p99 Error rate
8 virtual users
Add to cart 6.49 504 ms 586 ms 606 ms 0%
Cart 6.49 408 ms 479 ms 498 ms 0%
Checkout 6.49 315 ms 376 ms 392 ms 0%
32 virtual users
Add to cart 6.54 1.68 s 2 s 2.1 s 0%
Cart 6.54 1.77 s 1.93 s 1.99 s 0%
Checkout 6.54 1.4 s 1.61 s 1.71 s 0%

k6 with a fixed number of virtual users, 61 to 67 s per level. The error rate counts responses that weren't a 200.

Discourse

Signed in

Step Requests/s Median p95 p99 Error rate
1 virtual user
Topic page 3.55 125 ms 140 ms 310 ms 0%
Topic JSON 1.79 63.3 ms 71 ms 130 ms 0%
Home page 1.79 111 ms 171 ms 314 ms 0%
Categories page 0.897 131 ms 142 ms 154 ms 0%
User activity page 0.897 76.4 ms 83.3 ms 128 ms 0%
4 virtual users
Topic page 6.37 287 ms 361 ms 586 ms 0%
Topic JSON 3.18 144 ms 184 ms 197 ms 0%
Home page 3.14 255 ms 328 ms 643 ms 0%
Categories page 1.61 293 ms 370 ms 521 ms 0%
User activity page 1.59 176 ms 313 ms 459 ms 0%
16 virtual users
Topic page 6.11 1.05 s 1.26 s 1.5 s 0%
Topic JSON 3.07 916 ms 1.08 s 1.2 s 0%
Home page 3.05 1.03 s 1.2 s 1.37 s 0%
Categories page 1.52 1.05 s 1.22 s 1.29 s 0%
User activity page 1.55 958 ms 1.15 s 1.3 s 0%

Anonymous

Step Requests/s Median p95 p99 Error rate
16 virtual users
Topic page 127 34.5 ms 89.1 ms 115 ms 0%
Topic JSON 63.3 32.5 ms 86.9 ms 114 ms 0%
Home page 63.3 34.3 ms 87.6 ms 111 ms 0%
Categories page 31.7 32.9 ms 91.1 ms 120 ms 0%
User activity page 31.7 116 ms 182 ms 256 ms 0%

k6 with a fixed number of virtual users, 60 to 61 s per level, after a 21 s warmup. The error rate counts responses that weren't a 200. Discourse ran with 4 app workers and 1024 MB of Postgres shared buffers, what its installer picks for a 4 GB box. It added a 2 GB swapfile first, the way its installer does, and didn't need it. The container peaked at 2,267 MB.

Ghost

Readers

Step Requests/s Median p95 p99 Error rate
1 virtual user
Home page 14.1 21.2 ms 23.3 ms 30.7 ms 0%
Post page 35.2 10.3 ms 12.3 ms 18.6 ms 0%
Tag page 7.03 13.5 ms 15.2 ms 20.1 ms 0%
Author page 7.03 14.2 ms 15.8 ms 18.8 ms 0%
RSS feed 7.03 17.6 ms 20 ms 26.7 ms 0%
4 virtual users
Home page 17 62.7 ms 79.2 ms 105 ms 0%
Post page 42.6 42.9 ms 59 ms 71 ms 0%
Tag page 8.52 38.5 ms 54.8 ms 65.9 ms 0%
Author page 8.52 40.8 ms 56.3 ms 63.1 ms 0%
RSS feed 8.51 43.3 ms 59.6 ms 65.7 ms 0%
16 virtual users
Home page 17 237 ms 292 ms 314 ms 0%
Post page 42.6 196 ms 247 ms 271 ms 0%
Tag page 8.49 129 ms 169 ms 195 ms 0%
Author page 8.52 129 ms 168 ms 198 ms 0%
RSS feed 8.49 134 ms 174 ms 198 ms 0%

Staff in Ghost Admin

Step Requests/s Median p95 p99 Error rate
1 virtual user
Post list 7.83 19.7 ms 23.5 ms 29.5 ms 0%
Post editor 23.5 7.43 ms 9.94 ms 15.8 ms 0%
Saving a draft 23.5 18.5 ms 23.2 ms 29.9 ms 0%
Member list 7.83 22 ms 27 ms 32.7 ms 0%
4 virtual users
Post list 9.81 26.2 ms 30.9 ms 37.8 ms 0%
Post editor 29.5 13 ms 16.7 ms 20.6 ms 0%
Saving a draft 29.5 99.6 ms 124 ms 167 ms 0%
Member list 9.82 28.7 ms 33.9 ms 51.6 ms 0%

k6 with a fixed number of virtual users, 60 s per level, after a 20 s warmup. The error rate counts responses that weren't a 200. Ghost ran the way its ghost-docker setup runs it: one Node process, and MySQL with its stock 128 MB buffer pool. Memory peaked with Ghost at 333 MB and MySQL at 622 MB.

Wagtail

Visitors

Step Requests/s Median p95 p99 Error rate
1 virtual user
Home page 2.59 51.7 ms 54 ms 55.5 ms 0%
Blog index 2.59 173 ms 182 ms 184 ms 0%
Blog post 7.74 16.6 ms 25 ms 27.4 ms 0%
Bread page 7.76 18.3 ms 19.5 ms 20.4 ms 0%
Location page 2.59 17.3 ms 19.2 ms 20.5 ms 0%
Site search 2.57 28.8 ms 47.2 ms 48.2 ms 0%
4 virtual users
Home page 3.67 151 ms 202 ms 222 ms 0%
Blog index 3.68 548 ms 693 ms 738 ms 0%
Blog post 11 42 ms 72.9 ms 92.9 ms 0%
Bread page 11 46.7 ms 66.3 ms 79.1 ms 0%
Location page 3.7 45.7 ms 66.6 ms 78.2 ms 0%
Site search 3.65 69 ms 129 ms 155 ms 0%
16 virtual users
Home page 3.57 540 ms 742 ms 876 ms 0%
Blog index 3.6 1.34 s 1.57 s 1.65 s 0%
Blog post 10.7 290 ms 487 ms 578 ms 0%
Bread page 10.7 290 ms 486 ms 585 ms 0%
Location page 3.58 289 ms 456 ms 566 ms 0%
Site search 3.55 331 ms 511 ms 625 ms 0%

Editors in the Wagtail admin

Step Requests/s Median p95 p99 Error rate
1 virtual user
Admin dashboard 1.06 53.3 ms 58.2 ms 99.9 ms 0%
Page explorer 2.13 228 ms 246 ms 333 ms 0%
Page editor 3.18 107 ms 115 ms 191 ms 0%
Image library 1.06 41.7 ms 47.5 ms 331 ms 0%
Admin page search 1.06 21.9 ms 48.4 ms 77.7 ms 0%
4 virtual users
Admin dashboard 1.65 165 ms 355 ms 431 ms 0%
Page explorer 3.29 553 ms 936 ms 1 s 0%
Page editor 4.94 261 ms 524 ms 626 ms 0%
Image library 1.65 163 ms 368 ms 425 ms 0%
Admin page search 1.65 57.9 ms 155 ms 262 ms 0%

k6 with a fixed number of virtual users, 60 to 61 s per level, after a 20 s warmup. The error rate counts responses that weren't a 200. Wagtail ran the way bakerydemo ships it: uWSGI with 2 workers of 4 threads each, whatever the box, and Postgres and Redis on stock settings. Memory peaked with Wagtail at 302 MB, Postgres at 193 MB, and Redis at 26 MB.

Advertised vs. measured

Advertised Measured
vCPU 2 2
CPU Dedicated vCPU 0% of busy CPU time stolen
RAM 4 GB 3.7 GiB
Storage 50 GB NVMe 48.4 GiB
Architecture x86_64 x86_64
Virtualization KVM KVM

Guests usually report slightly less RAM and disk than advertised because of kernel reservations and filesystem overhead.

Detected hardware

CPU model
AMD EPYC-Milan Processor
CPU cores
2
CPU frequency
2400 MHz
AES-NI
Yes
VM-x/AMD-V
Yes
RAM reported
3.7 GiB
Swap reported
0 MiB
Disk reported
48.4 GiB
Virtualization reported
KVM
CPU steal
0% of busy CPU time stolen
Distro
Ubuntu 24.04.4 LTS
Kernel
6.8.0-136-generic
Architecture
x86_64
IPv4 reachable
Yes
IPv6 reachable
No

Advertised at test time

Plan
C3-4
Product line
Compute Optimized C3
Location
Milan
vCPU
2 dedicated vCPU
Architecture
x86_64
RAM
4 GB
Storage
50 GB NVMe
Virtualization
KVM
Network
250 Mbps
Transfer
Unmetered
Port 25
Unknown
Monthly price at test time
$39.42
Hourly price at test time
$0.054
YABS version
v2026-09-20
Ran at
Oct 12, 2026, 02:32:08 UTC
Parser version
5

Run timeline

Queued
Oct 12, 2026, 01:06:45 UTC
First update from the server
Oct 12, 2026, 02:31:43 UTC
Completed
Oct 12, 2026, 04:06:00 UTC
Server deleted
Oct 12, 2026, 04:06:00 UTC
Cost of this run
$0.086
Watch party 0 watching

This party closed at . It's read only now.

Watch party

Loading the party