Skip to main content

Hetzner CPX22

Nuremberg DC Park 1

Benchmark complete
Finished Oct 12, 2026
Geekbench 7 single-core
1,699
Faster than 86% of 2 vCPU plans
Geekbench 7 multi-core
3,274
Faster than 91% of 2 vCPU plans
Disk 4K IOPS
74,046
fio random read and write, 50/50
Monthly price at test time
$22.99
142 multi-core points per dollar
Same price today

Server

  • Provider
    Hetzner
  • Plan
    CPX22
  • Location
    Nuremberg DC Park 1
  • vCPU
    2
  • RAM
    4 GB
  • Storage
    80 GB NVMe
  • Network
    Unknown
  • OS image
    ubuntu-24.04
  • Trigger
    Manual
  • Queued at
    Oct 12, 2026, 01:06:45 UTC
Visit Hetzner

Latest runs of this plan

Full history of this plan

CPU

Version Single-core Multi-core
Geekbench 7 1,699 3,274 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
1,735
24th of 106
2,995
35th of 90
Navigation
1,779
9th of 106
Single-core only
HTML5 Browser
1,638
22nd of 106
Single-core only
PDF Viewer
1,756
21st of 106
Single-core only
Photo Library
1,702
11th of 106
3,405
32nd of 90
Clang
1,556
19th of 106
3,194
39th of 90
Text Processing
1,544
23rd of 106
2,906
39th of 90
Asset Compression
1,710
22nd of 106
3,381
39th of 90
HDR
1,770
20th of 106
3,306
28th of 90
Photo Editor
1,994
19th of 106
3,970
25th of 90
Ray Tracer
1,636
22nd of 106
3,141
43rd of 90
Structure from Motion
1,739
27th of 106
Single-core only
Game Physics
1,545
27th of 106
Single-core only
Video Encoder
1,665
23rd of 106
Single-core only
Audio Encoder
1,734
23rd of 106
Single-core only
Video Player
1,729
32nd 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 49.6 GB/s 3 3.25%
Scale higher is better 41.3 GB/s 3 2.62%
Add higher is better 38.6 GB/s 3 4.2%
Triad higher is better 39 GB/s 3 4.1%

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 1,306/s 3 0.702%
RSA 4096 verifies higher is better 31,756/s 3 0.648%
SHA-256 higher is better 1.96 GB/s 3 0.81%
AES-256-GCM higher is better 19.6 GB/s 3 0.896%
ChaCha20-Poly1305 higher is better 7.36 GB/s 3 0.399%

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 4.2 ms 3 0.22%
Requests per second, 90th percentile higher is better 470/s 3 0%
Request latency, 99.9th percentile lower is better 8.92 ms 3 8.35%
Requests per second, average higher is better 460/s 3 0.229%

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

Disk

Block size Read Write Total
4K 144.4 MB/s (36,974 IOPS) 144.8 MB/s (37,072 IOPS) 289.2 MB/s (74,046 IOPS)
64K 1.58 GB/s (25,917 IOPS) 1.59 GB/s (26,053 IOPS) 3.17 GB/s (51,970 IOPS)
512K 2.31 GB/s (4,733 IOPS) 2.43 GB/s (4,984 IOPS) 4.74 GB/s (9,717 IOPS)
1M 2.69 GB/s (2,756 IOPS) 2.87 GB/s (2,940 IOPS) 5.56 GB/s (5,696 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 1.84 Gbps 5.36 Gbps
Eranium, Amsterdam, NL (100G) IPv4 2.71 Gbps 4.64 Gbps
Leaseweb, Singapore, SG (10G) IPv4 1.15 Gbps 1.4 Gbps
Clouvider, Los Angeles, CA, US (10G) IPv4 1.07 Gbps 1.22 Gbps
Leaseweb, NYC, NY, US (10G) IPv4 2.06 Gbps 2.54 Gbps

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 3,358 0.298 ms
fdatasync 3,477 0.288 ms
fsync 1,669 0.599 ms
fsync_writethrough Not supported here n/a
open_sync 1,922 0.52 ms
Two 8 kB writes
open_datasync 1,709 0.585 ms
fdatasync 3,054 0.327 ms
fsync 1,728 0.579 ms
fsync_writethrough Not supported here n/a
open_sync 986 1.01 ms
open_sync at different write sizes
1 * 16kB open_sync write 1,861 0.537 ms
2 * 8kB open_sync writes 1,054 0.949 ms
4 * 4kB open_sync writes 410 2.44 ms
8 * 2kB open_sync writes 263 3.81 ms
16 * 1kB open_sync writes 117 8.55 ms
fsync on a non-write descriptor
write, fsync, close 1,316 0.76 ms
write, close, fsync 1,615 0.619 ms
Non-synced 8 kB writes
write 608,575 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 1,145 0.784 ms 1.36 ms 1.82 ms 4.84 ms 0
Read/write (TPC-B like) 8 4,869 1.46 ms 2.9 ms 4.48 ms 8.46 ms 0
Read/write (TPC-B like) 32 4,749 4.96 ms 17.4 ms 30.1 ms 50.1 ms 0
Read only 1 23,940 0.038 ms 0.06 ms 0.084 ms 0.189 ms 0
Read only 8 62,775 0.109 ms 0.189 ms 0.243 ms 0.418 ms 0
Read only 32 64,856 0.375 ms 0.705 ms 1 ms 1.53 ms 0

Scale 10 (1,000,000 accounts), loaded in 1.11 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 1,030 0.857 ms 1.57 ms 2.23 ms 4.88 ms 0
Read/write (TPC-B like) 8 4,437 1.68 ms 2.99 ms 4.47 ms 7.67 ms 0
Read/write (TPC-B like) 32 5,708 5.22 ms 8.9 ms 12.3 ms 18.5 ms 0
Read only 1 18,771 0.047 ms 0.077 ms 0.143 ms 0.548 ms 0
Read only 8 54,337 0.131 ms 0.222 ms 0.287 ms 0.674 ms 0
Read only 32 50,152 0.535 ms 0.886 ms 1.2 ms 2.15 ms 0

Scale 100 (10,000,000 accounts), loaded in 10.9 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 1 318 6,351 2.86 ms 4.33 ms 9.39 ms 0
Read/write 8 959 19,188 8.13 ms 13 ms 17.3 ms 0
Read/write 32 1,065 21,301 29.7 ms 47.5 ms 58.9 ms 0
Read only 1 976 15,615 0.954 ms 1.39 ms 1.86 ms 0
Read only 8 2,220 35,515 3.49 ms 5.47 ms 6.43 ms 0
Read only 32 2,134 34,141 14.2 ms 20.4 ms 23.9 ms 0

10 tables of 100,000 rows, loaded in 6.8 s. 60 s per run, one connection per thread.

Redis

memtier

Mix Operations/s Hits/s Median p99 p99.9
1:10 set:get 66,818 60,744 0.599 ms 2.93 ms 4.42 ms
1:1 set:get 64,538 32,269 0.687 ms 2.96 ms 4.35 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 801 2,953 4,283 440 ms
Express express-postgresjs 8,827 12,704 44,629 60.5 ms
Fastify fastify-postgres 8,344 12,836 48,607 63.9 ms
NestJS nestjs 6,329 6,942 28,750 76.1 ms
Nuxt nuxt 7,592 12,391 41,655 72.1 ms
FastAPI fastapi 4,541 5,768 26,793 290 ms
Django django-postgresql 2,258 2,537 6,409 224 ms
Rails rails 3,835 5,175 18,776 93.8 ms
Laravel laravel 1,032 1,055 2,002 293 ms
Symfony symfony 1,233 1,394 3,982 242 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 2.9 550 ms 652 ms 677 ms 0%
Shop page 2.9 743 ms 909 ms 949 ms 0%
Category page 2.9 467 ms 803 ms 874 ms 0%
Product page 2.9 503 ms 631 ms 665 ms 0%
Product search 2.9 441 ms 551 ms 586 ms 0%
32 virtual users
Home page 2.87 2.04 s 2.34 s 2.56 s 0%
Shop page 2.87 2.71 s 3.02 s 3.16 s 0%
Category page 2.87 2.34 s 2.79 s 2.91 s 0%
Product page 2.87 2.05 s 2.21 s 2.37 s 0%
Product search 2.87 1.9 s 2.08 s 2.15 s 0%

Cart and checkout

Step Requests/s Median p95 p99 Error rate
8 virtual users
Add to cart 5.63 588 ms 682 ms 728 ms 0%
Cart 5.63 477 ms 565 ms 581 ms 0%
Checkout 5.63 374 ms 450 ms 481 ms 0%
32 virtual users
Add to cart 5.62 1.97 s 2.38 s 2.46 s 0%
Cart 5.62 2.05 s 2.3 s 2.37 s 0%
Checkout 5.62 1.63 s 1.87 s 1.95 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 2.39 192 ms 234 ms 316 ms 0%
Topic JSON 1.19 92.7 ms 117 ms 250 ms 0%
Home page 1.19 170 ms 204 ms 368 ms 0%
Categories page 0.597 198 ms 233 ms 320 ms 0%
User activity page 0.597 118 ms 195 ms 333 ms 0%
4 virtual users
Topic page 4.96 365 ms 501 ms 793 ms 0%
Topic JSON 2.47 184 ms 250 ms 276 ms 0%
Home page 2.45 322 ms 406 ms 549 ms 0%
Categories page 1.27 382 ms 493 ms 725 ms 0%
User activity page 1.22 232 ms 313 ms 763 ms 0%
16 virtual users
Topic page 4.9 1.31 s 1.55 s 1.73 s 0%
Topic JSON 2.45 1.13 s 1.33 s 1.51 s 0%
Home page 2.47 1.27 s 1.45 s 1.66 s 0%
Categories page 1.22 1.29 s 1.51 s 1.76 s 0%
User activity page 1.23 1.21 s 1.37 s 1.49 s 0%

Anonymous

Step Requests/s Median p95 p99 Error rate
16 virtual users
Topic page 106 44.2 ms 101 ms 136 ms 0%
Topic JSON 52.9 41.7 ms 99.8 ms 136 ms 0%
Home page 52.9 44 ms 101 ms 145 ms 0%
Categories page 26.5 43.9 ms 99.5 ms 130 ms 0%
User activity page 26.4 133 ms 209 ms 291 ms 0%

k6 with a fixed number of virtual users, 60 to 62 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 768 MB of Postgres shared buffers, what its installer picks for a 3 GB box. It added a 2 GB swapfile first, the way its installer does, and didn't need it. The container peaked at 2,132 MB.

Ghost

Readers

Step Requests/s Median p95 p99 Error rate
1 virtual user
Home page 10.8 25.4 ms 32.4 ms 40.5 ms 0%
Post page 26.9 14 ms 20.2 ms 27.5 ms 0%
Tag page 5.38 17.4 ms 22.5 ms 30.9 ms 0%
Author page 5.38 17.2 ms 25.3 ms 32.5 ms 0%
RSS feed 5.36 22.1 ms 28.2 ms 39.7 ms 0%
4 virtual users
Home page 13.4 76.8 ms 104 ms 125 ms 0%
Post page 33.4 55.5 ms 78.7 ms 90.8 ms 0%
Tag page 6.69 49.1 ms 68.9 ms 76.3 ms 0%
Author page 6.66 50.4 ms 72.2 ms 90.9 ms 0%
RSS feed 6.66 53.7 ms 75.2 ms 82.6 ms 0%
16 virtual users
Home page 12.5 324 ms 392 ms 417 ms 0%
Post page 31.4 267 ms 335 ms 370 ms 0%
Tag page 6.28 180 ms 228 ms 276 ms 0%
Author page 6.29 178 ms 233 ms 269 ms 0%
RSS feed 6.28 182 ms 234 ms 266 ms 0%

Staff in Ghost Admin

Step Requests/s Median p95 p99 Error rate
1 virtual user
Post list 6.05 23.9 ms 30.2 ms 34.5 ms 0%
Post editor 18.2 9.52 ms 13.3 ms 17.3 ms 0%
Saving a draft 18.2 24.8 ms 31.3 ms 42.2 ms 0%
Member list 6.06 27.4 ms 37.3 ms 42.2 ms 0%
4 virtual users
Post list 7.91 31.3 ms 41.1 ms 50.8 ms 0%
Post editor 23.8 14.8 ms 20.3 ms 26.7 ms 0%
Saving a draft 23.8 126 ms 155 ms 201 ms 0%
Member list 7.93 31 ms 40.5 ms 43.1 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 317 MB and MySQL at 617 MB.

Wagtail

Visitors

Step Requests/s Median p95 p99 Error rate
1 virtual user
Home page 1.84 68.3 ms 88.2 ms 96 ms 0%
Blog index 1.84 240 ms 282 ms 310 ms 0%
Blog post 5.49 23.3 ms 39.7 ms 43.9 ms 0%
Bread page 5.51 24.6 ms 41.9 ms 44.8 ms 0%
Location page 1.84 23.1 ms 39.9 ms 42 ms 0%
Site search 1.82 40.1 ms 68.4 ms 80.8 ms 0%
4 virtual users
Home page 3.57 148 ms 228 ms 255 ms 0%
Blog index 3.6 499 ms 765 ms 834 ms 0%
Blog post 10.7 45.2 ms 88.2 ms 124 ms 0%
Bread page 10.7 50.3 ms 79.1 ms 92.7 ms 0%
Location page 3.6 48.8 ms 75.5 ms 81.2 ms 0%
Site search 3.55 77.8 ms 133 ms 179 ms 0%
16 virtual users
Home page 3.92 484 ms 657 ms 750 ms 0%
Blog index 3.98 1.12 s 1.36 s 1.49 s 0%
Blog post 11.8 270 ms 423 ms 521 ms 0%
Bread page 11.8 281 ms 446 ms 551 ms 0%
Location page 3.95 281 ms 431 ms 513 ms 0%
Site search 3.92 326 ms 508 ms 547 ms 0%

Editors in the Wagtail admin

Step Requests/s Median p95 p99 Error rate
1 virtual user
Admin dashboard 0.828 71.8 ms 100 ms 105 ms 0%
Page explorer 1.66 281 ms 318 ms 357 ms 0%
Page editor 2.49 139 ms 186 ms 286 ms 0%
Image library 0.828 54.7 ms 84.4 ms 454 ms 0%
Admin page search 0.828 29.5 ms 65.4 ms 74.8 ms 0%
4 virtual users
Admin dashboard 1.45 199 ms 381 ms 485 ms 0%
Page explorer 2.93 607 ms 996 ms 1.06 s 0%
Page editor 4.36 291 ms 589 ms 651 ms 0%
Image library 1.46 182 ms 403 ms 461 ms 0%
Admin page search 1.45 67.6 ms 221 ms 335 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 301 MB, Postgres at 163 MB, and Redis at 26 MB.

Advertised vs. measured

Advertised Measured
vCPU 2 2
CPU Shared vCPU 0% of busy CPU time stolen
RAM 4 GB 3.7 GiB
Storage 80 GB NVMe 75 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-Genoa Processor
CPU cores
2
CPU frequency
2396 MHz
AES-NI
Yes
VM-x/AMD-V
No
RAM reported
3.7 GiB
Swap reported
0 MiB
Disk reported
75 GiB
Virtualization reported
KVM
CPU steal
0% of busy CPU time stolen
Distro
Ubuntu 24.04.4 LTS
Kernel
6.8.0-138-generic
Architecture
x86_64
IPv4 reachable
Yes
IPv6 reachable
Yes

Advertised at test time

Plan
CPX22
Product line
regular_purpose
Location
Nuremberg DC Park 1
vCPU
2 shared vCPU
Architecture
x86_64
RAM
4 GB
Storage
80 GB NVMe
Virtualization
KVM
Network
Unknown
Transfer
21.99 TB
Port 25
On request
Monthly price at test time
$22.99
Hourly price at test time
$0.037
YABS version
v2026-09-20
Ran at
Oct 12, 2026, 04:01:51 UTC
Parser version
5

Run timeline

Queued
Oct 12, 2026, 01:06:45 UTC
First update from the server
Oct 12, 2026, 04:00:40 UTC
Completed
Oct 12, 2026, 05:34:42 UTC
Server deleted
Oct 12, 2026, 05:34:42 UTC
Cost of this run
$0.058
Watch party 0 watching

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

Watch party

Loading the party