Hetzner CPX22

Falkenstein DC Park 1

Benchmark complete

Finished Oct 10, 2026

Geekbench 7 single-core

1,886

Faster than 95% of 2 vCPU plans

Geekbench 7 multi-core

3,588

Faster than 94% of 2 vCPU plans

Disk 4K IOPS

137,643

fio random read and write, 50/50

Monthly price at test time

$22.99

156 multi-core points per dollar

Same price today

Server

Hetzner CPX22

Provider

Hetzner
Plan

CPX22
Location

Falkenstein DC Park 1
vCPU

2
RAM

4 GB
Storage

80 GB NVMe
Network

Unknown
OS image

ubuntu-24.04
Trigger

Scheduled
Queued at

Oct 10, 2026, 17:02:00 UTC

Visit Hetzner

Latest runs of this plan

Helsinki DC Park 1 Oct 11, 2026 1,674

Full history of this plan

CPU

Version	Single-core	Multi-core	Result link
Geekbench 7	1,886	3,588	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	1,804 8th of 70	3,077 23rd of 62
Navigation	1,602 22nd of 70	Single-core only
HTML5 Browser	1,882 8th of 70	Single-core only
PDF Viewer	1,994 10th of 70	Single-core only
Photo Library	1,903 4th of 70	3,736 21st of 62
Clang	1,638 10th of 70	3,221 28th of 62
Text Processing	1,790 9th of 70	3,228 27th of 62
Asset Compression	1,885 9th of 70	3,721 27th of 62
HDR	1,985 4th of 70	3,420 19th of 62
Photo Editor	2,719 4th of 70	5,357 9th of 62
Ray Tracer	1,589 20th of 70	3,371 29th of 62
Structure from Motion	1,833 10th of 70	Single-core only
Game Physics	1,738 9th of 70	Single-core only
Video Encoder	1,897 9th of 70	Single-core only
Audio Encoder	1,845 6th of 70	Single-core only
Video Player	2,347 3rd of 70	Single-core only

Rank among the 70 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	65.5 GB/s	3	25.1%
Scale higher is better	60.3 GB/s	3	31.8%
Add higher is better	61.5 GB/s	3	31.6%
Triad higher is better	61.9 GB/s	3	32.6%

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	1,294/s	3	1.33%
RSA 4096 verifies higher is better	35,277/s	3	0.833%
SHA-256 higher is better	1.96 GB/s	3	1.09%
AES-256-GCM higher is better	34 GB/s	3	1.39%
ChaCha20-Poly1305 higher is better	8.63 GB/s	3	0.249%

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	3.64 ms	3	0.22%
Requests per second, 90th percentile higher is better	508/s	3	1.38%
Request latency, 99.9th percentile lower is better	11.3 ms	3	1.14%
Requests per second, average higher is better	494/s	3	0.744%

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

Disk

Block size	Read	Write	Total
4K	268.5 MB/s (68,731 IOPS)	269.2 MB/s (68,912 IOPS)	537.7 MB/s (137,643 IOPS)
64K	2.16 GB/s (35,331 IOPS)	2.17 GB/s (35,517 IOPS)	4.32 GB/s (70,848 IOPS)
512K	2.36 GB/s (4,830 IOPS)	2.48 GB/s (5,087 IOPS)	4.84 GB/s (9,917 IOPS)
1M	2.35 GB/s (2,405 IOPS)	2.51 GB/s (2,565 IOPS)	4.85 GB/s (4,970 IOPS)

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

Network

IPv4 transfer speed

Target	Protocol	Send	Receive
Clouvider, London, UK (10G)	IPv4	6.87 Gbps	4.38 Gbps
Eranium, Amsterdam, NL (100G)	IPv4	2.21 Gbps	5.5 Gbps
Leaseweb, Singapore, SG (10G)	IPv4	1.12 Gbps	1.11 Gbps
Clouvider, Los Angeles, CA, US (10G)	IPv4	804 Mbps	1.01 Gbps
Leaseweb, NYC, NY, US (10G)	IPv4	1.61 Gbps	2.03 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	4,141	0.241 ms
fdatasync	3,750	0.267 ms
fsync	2,298	0.435 ms
fsync_writethrough	Not supported here	n/a
open_sync	2,370	0.422 ms
Two 8 kB writes
open_datasync	2,093	0.478 ms
fdatasync	3,681	0.272 ms
fsync	2,184	0.458 ms
fsync_writethrough	Not supported here	n/a
open_sync	1,181	0.847 ms
open_sync at different write sizes
1 * 16kB open_sync write	2,288	0.437 ms
2 * 8kB open_sync writes	1,228	0.814 ms
4 * 4kB open_sync writes	613	1.63 ms
8 * 2kB open_sync writes	314	3.19 ms
16 * 1kB open_sync writes	151	6.63 ms
fsync on a non-write descriptor
write, fsync, close	2,069	0.483 ms
write, close, fsync	1,976	0.506 ms
Non-synced 8 kB writes
write	682,965	0.001 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
Read/write (TPC-B like)	1	1,176	0.837 ms	1.13 ms	1.52 ms	3.93 ms
Read/write (TPC-B like)	8	5,315	1.3 ms	2.88 ms	4.16 ms	6.84 ms
Read/write (TPC-B like)	32	5,481	4.39 ms	14.5 ms	25.2 ms	43.8 ms
Read only	1	23,599	0.034 ms	0.073 ms	0.095 ms	0.188 ms
Read only	8	83,278	0.082 ms	0.135 ms	0.183 ms	0.331 ms
Read only	32	72,380	0.329 ms	0.693 ms	1.02 ms	1.63 ms

Scale 10 (1,000,000 accounts), loaded in 0.98 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
Read/write (TPC-B like)	1	1,034	0.937 ms	1.32 ms	1.78 ms	3.41 ms
Read/write (TPC-B like)	8	4,918	1.43 ms	3.02 ms	4.53 ms	7.68 ms
Read/write (TPC-B like)	32	7,194	3.98 ms	7.79 ms	11 ms	16.6 ms
Read only	1	20,031	0.039 ms	0.084 ms	0.157 ms	0.617 ms
Read only	8	72,510	0.098 ms	0.162 ms	0.218 ms	0.576 ms
Read only	32	63,962	0.415 ms	0.687 ms	0.999 ms	2.1 ms

Scale 100 (10,000,000 accounts), loaded in 10.1 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
Read/write	1	276	5,527	3.55 ms	5.18 ms	7.99 ms
Read/write	8	1,103	22,064	7.04 ms	11.9 ms	15.8 ms
Read/write	32	1,349	26,985	23.5 ms	38.2 ms	48.3 ms
Read only	1	1,002	16,024	0.872 ms	1.67 ms	2.03 ms
Read only	8	2,971	47,529	2.62 ms	3.19 ms	4.25 ms
Read only	32	2,871	45,943	10.8 ms	13.7 ms	15.8 ms

10 tables of 100,000 rows, loaded in 6.1 s. 60 s per run, one connection per thread.

Redis

memtier

Mix	Operations/s	Hits/s	Median	p99	p99.9
1:10 set:get	67,746	61,587	0.599 ms	3.06 ms	4.42 ms
1:1 set:get	67,333	33,666	0.615 ms	3.04 ms	4.42 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	895	3,081	4,339	437 ms
Express express-postgresjs	9,522	14,231	54,448	48.4 ms
Fastify fastify-postgres	9,068	14,601	57,284	61.1 ms
NestJS nestjs	8,360	9,603	43,132	65.7 ms
Nuxt nuxt	8,245	14,429	51,109	66.8 ms
FastAPI fastapi	5,811	6,809	28,878	153 ms
Django django-postgresql	2,603	2,742	7,404	143 ms
Rails rails	4,264	6,282	25,846	111 ms
Laravel laravel	1,029	1,046	2,087	284 ms
Symfony symfony	1,256	1,374	4,445	226 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	2.83	554 ms	647 ms	677 ms	0%
Shop page	2.83	757 ms	917 ms	942 ms	0%
Category page	2.83	482 ms	750 ms	807 ms	0%
Product page	2.83	516 ms	635 ms	675 ms	0%
Product search	2.83	448 ms	539 ms	567 ms	0%
32 virtual users
Home page	2.89	2.04 s	2.3 s	2.47 s	0%
Shop page	2.89	2.66 s	3.06 s	3.16 s	0%
Category page	2.89	2.33 s	2.78 s	2.97 s	0%
Product page	2.89	2.02 s	2.28 s	2.41 s	0%
Product search	2.89	1.92 s	2.11 s	2.17 s	0%

Cart and checkout

Step	Requests/s	Median	p95	p99	Error rate
8 virtual users
Add to cart	5.64	597 ms	709 ms	741 ms	0%
Cart	5.64	488 ms	577 ms	612 ms	0%
Checkout	5.64	372 ms	448 ms	472 ms	0%
32 virtual users
Add to cart	5.56	1.97 s	2.41 s	2.55 s	0%
Cart	5.56	2.04 s	2.29 s	2.38 s	0%
Checkout	5.56	1.66 s	1.98 s	2.08 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.

Advertised vs. measured

Spec	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: 2200 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: Falkenstein 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 10, 2026, 17:02:39 UTC
Parser version: 5

Run timeline

Queued: Oct 10, 2026, 17:02:00 UTC
Server created: Oct 10, 2026, 17:02:33 UTC
First update from the server: Oct 10, 2026, 17:02:33 UTC
Completed: Oct 10, 2026, 18:16:22 UTC
Server deleted: Oct 10, 2026, 18:16:22 UTC
Cost of this run: $0.046

TestResultTimevs last run

Watch party 0 watching

This party closed at 18:12 UTC. It's read only now.

Loading the party

RackCensus

Server

Latest runs of this plan

CPU

Geekbench 7 workloads

Memory, TLS and scheduler

Memory bandwidth (STREAM)

TLS and crypto (OpenSSL)

Scheduler latency (schbench)

Disk

Network

IPv4 transfer speed

Datastores

PostgreSQL

fsync speed (pg_test_fsync)

pgbench at scale 10

pgbench at scale 100

MySQL

sysbench OLTP

Redis

memtier

Web frameworks

WooCommerce

Page mix

Cart and checkout

Advertised vs. measured

Detected hardware

Advertised at test time

Run timeline

Watch party