Skip to main content

Hetzner CPX22

Helsinki DC Park 1

Benchmark complete
Finished Oct 11, 2026
Geekbench 7 single-core
1,674
Faster than 85% of 2 vCPU plans
Geekbench 7 multi-core
3,257
Faster than 88% of 2 vCPU plans
Disk 4K IOPS
79,395
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
    Helsinki DC Park 1
  • vCPU
    2
  • RAM
    4 GB
  • Storage
    80 GB NVMe
  • Network
    Unknown
  • OS image
    ubuntu-24.04
  • Trigger
    Scheduled
  • Queued at
    Oct 11, 2026, 09:32:00 UTC
Visit Hetzner

Latest runs of this plan

Full history of this plan

CPU

Version Single-core Multi-core
Geekbench 7 1,674 3,257 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
1,747
18th of 71
3,038
24th of 63
Navigation
1,582
23rd of 71
Single-core only
HTML5 Browser
1,653
15th of 71
Single-core only
PDF Viewer
1,652
19th of 71
Single-core only
Photo Library
1,767
5th of 71
3,345
24th of 63
Clang
1,475
16th of 71
3,095
30th of 63
Text Processing
1,609
13th of 71
2,884
30th of 63
Asset Compression
1,710
16th of 71
3,370
30th of 63
HDR
1,752
15th of 71
3,291
21st of 63
Photo Editor
2,024
13th of 71
3,931
18th of 63
Ray Tracer
1,597
18th of 71
3,205
31st of 63
Structure from Motion
1,682
24th of 71
Single-core only
Game Physics
1,611
17th of 71
Single-core only
Video Encoder
1,611
18th of 71
Single-core only
Audio Encoder
1,591
24th of 71
Single-core only
Video Player
1,780
18th of 71
Single-core only

Rank among the 71 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 49.7 GB/s 3 1.72%
Scale higher is better 41.2 GB/s 3 0.545%
Add higher is better 39 GB/s 3 1.12%
Triad higher is better 38.9 GB/s 3 1.14%

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 0.534%
RSA 4096 verifies higher is better 31,161/s 3 3.02%
SHA-256 higher is better 1.95 GB/s 3 1.16%
AES-256-GCM higher is better 19.5 GB/s 3 0.863%
ChaCha20-Poly1305 higher is better 7.37 GB/s 3 0.376%

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 4.12 ms 3 0.388%
Requests per second, 90th percentile higher is better 478/s 3 0.362%
Request latency, 99.9th percentile lower is better 9.46 ms 3 8.9%
Requests per second, average higher is better 465/s 3 1.1%

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

Disk

Block size Read Write Total
4K 154.9 MB/s (39,645 IOPS) 155.3 MB/s (39,750 IOPS) 310.1 MB/s (79,395 IOPS)
64K 1.55 GB/s (25,374 IOPS) 1.56 GB/s (25,507 IOPS) 3.11 GB/s (50,881 IOPS)
512K 2.47 GB/s (5,057 IOPS) 2.6 GB/s (5,325 IOPS) 5.07 GB/s (10,382 IOPS)
1M 2.59 GB/s (2,653 IOPS) 2.76 GB/s (2,829 IOPS) 5.35 GB/s (5,482 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 4.59 Gbps 5.56 Gbps
Eranium, Amsterdam, NL (100G) IPv4 2.16 Gbps 3.61 Gbps
Leaseweb, Singapore, SG (10G) IPv4 1.05 Gbps 1.26 Gbps
Clouvider, Los Angeles, CA, US (10G) IPv4 1 Gbps 1.19 Gbps
Leaseweb, NYC, NY, US (10G) IPv4 2.12 Gbps 2.36 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 3,809 0.263 ms
fdatasync 3,696 0.271 ms
fsync 1,929 0.518 ms
fsync_writethrough Not supported here n/a
open_sync 2,197 0.455 ms
Two 8 kB writes
open_datasync 2,038 0.491 ms
fdatasync 2,996 0.334 ms
fsync 1,777 0.563 ms
fsync_writethrough Not supported here n/a
open_sync 1,161 0.861 ms
open_sync at different write sizes
1 * 16kB open_sync write 2,168 0.461 ms
2 * 8kB open_sync writes 1,082 0.924 ms
4 * 4kB open_sync writes 549 1.82 ms
8 * 2kB open_sync writes 196 5.1 ms
16 * 1kB open_sync writes 116 8.62 ms
fsync on a non-write descriptor
write, fsync, close 1,609 0.622 ms
write, close, fsync 1,624 0.616 ms
Non-synced 8 kB writes
write 613,535 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 1,058 0.895 ms 1.4 ms 1.6 ms 3.21 ms 0
Read/write (TPC-B like) 8 5,012 1.42 ms 2.79 ms 4 ms 6.57 ms 0
Read/write (TPC-B like) 32 4,898 4.84 ms 16.8 ms 28.3 ms 49 ms 0
Read only 1 25,959 0.035 ms 0.053 ms 0.07 ms 0.145 ms 0
Read only 8 67,134 0.102 ms 0.17 ms 0.214 ms 0.341 ms 0
Read only 32 65,101 0.374 ms 0.713 ms 1.02 ms 1.54 ms 0

Scale 10 (1,000,000 accounts), loaded in 1.05 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) 1 869 1.16 ms 1.66 ms 2.22 ms 3.79 ms 0
Read/write (TPC-B like) 8 4,450 1.69 ms 2.94 ms 4.41 ms 7.18 ms 0
Read/write (TPC-B like) 32 5,475 5.49 ms 9.08 ms 12.1 ms 17.1 ms 0
Read only 1 20,299 0.043 ms 0.07 ms 0.117 ms 0.48 ms 0
Read only 8 52,167 0.137 ms 0.228 ms 0.289 ms 0.674 ms 0
Read only 32 53,221 0.5 ms 0.826 ms 1.15 ms 2.06 ms 0

Scale 100 (10,000,000 accounts), loaded in 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.

MySQL

sysbench OLTP

Workload Threads Transactions/s Queries/s Median p95 p99 Errors
Read/write 1 328 6,557 2.76 ms 4.41 ms 8.43 ms 0
Read/write 8 1,038 20,759 7.57 ms 11.4 ms 14.7 ms 0
Read/write 32 1,118 22,361 28.7 ms 45 ms 54.8 ms 0
Read only 1 989 15,819 0.954 ms 1.34 ms 1.64 ms 0
Read only 8 2,263 36,215 3.3 ms 5.57 ms 6.55 ms 0
Read only 32 2,213 35,402 14 ms 19.6 ms 22.7 ms 0

10 tables of 100,000 rows, loaded in 6.2 s. 60 s per run, one connection per thread.

Redis

memtier

Mix Operations/s Hits/s Median p99 p99.9
1:10 set:get 71,331 64,846 0.655 ms 2.66 ms 3.76 ms
1:1 set:get 61,553 30,776 0.751 ms 2.99 ms 4.29 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 835 2,867 4,299 427 ms
Express express-postgresjs 8,415 12,579 44,316 59.7 ms
Fastify fastify-postgres 8,159 13,363 50,919 57.5 ms
NestJS nestjs 6,294 6,991 28,767 83.2 ms
Nuxt nuxt 7,676 12,364 42,824 95.9 ms
FastAPI fastapi 4,715 5,828 26,726 225 ms
Django django-postgresql 2,194 2,517 6,270 190 ms
Rails rails 3,976 5,146 17,926 86.1 ms
Laravel laravel 967 1,036 2,000 301 ms
Symfony symfony 1,222 1,350 3,755 239 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 2.92 539 ms 639 ms 706 ms 0%
Shop page 2.92 728 ms 847 ms 904 ms 0%
Category page 2.92 465 ms 730 ms 801 ms 0%
Product page 2.92 502 ms 602 ms 673 ms 0%
Product search 2.92 435 ms 528 ms 568 ms 0%
32 virtual users
Home page 2.89 2.05 s 2.32 s 2.41 s 0%
Shop page 2.89 2.65 s 3.03 s 3.2 s 0%
Category page 2.89 2.34 s 2.79 s 2.93 s 0%
Product page 2.89 2.05 s 2.31 s 2.45 s 0%
Product search 2.89 1.88 s 2.14 s 2.2 s 0%

Cart and checkout

Step Requests/s Median p95 p99 Error rate
8 virtual users
Add to cart 5.45 603 ms 737 ms 790 ms 0%
Cart 5.45 480 ms 604 ms 631 ms 0%
Checkout 5.45 374 ms 466 ms 497 ms 0%
32 virtual users
Add to cart 5.58 1.96 s 2.37 s 2.47 s 0%
Cart 5.58 2.05 s 2.27 s 2.33 s 0%
Checkout 5.58 1.66 s 1.96 s 2.01 s 0%

k6 with a fixed number of virtual users, 61 to 66 s per level. The error rate counts responses that weren't a 200.

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
2400 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
Helsinki 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 11, 2026, 09:33:05 UTC
Parser version
5

Run timeline

Queued
Oct 11, 2026, 09:32:00 UTC
Server created
Oct 11, 2026, 09:32:39 UTC
First update from the server
Oct 11, 2026, 09:32:39 UTC
Completed
Oct 11, 2026, 10:47:09 UTC
Server deleted
Oct 11, 2026, 10:47:09 UTC
Cost of this run
$0.046
Watch party 0 watching

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

Watch party

Loading the party