Hetzner CPX11

Ashburn, VA

Benchmark complete

Finished Oct 10, 2026

Geekbench 7 single-core

1,300

Faster than 55% of 2 vCPU plans

Geekbench 7 multi-core

Skipped

Disk 4K IOPS

53,855

fio random read and write, 50/50

Monthly price at test time

$20.49

Same price today

Server

Hetzner CPX11

Provider

Hetzner
Plan

CPX11
Location

Ashburn, VA
vCPU

2
RAM

2 GB
Storage

40 GB NVMe
Network

Unknown
OS image

ubuntu-24.04
Trigger

Scheduled
Queued at

Oct 10, 2026, 17:02:00 UTC

Visit Hetzner

CPU

Version	Single-core	Multi-core	Result link
Geekbench 7	1,300	Skipped	Geekbench Browser

Multi-core was skipped because Geekbench 7 needs about 3.1 GB of free memory to run 2 threads, and this server didn't have that much. Single-core needs much less, so it still ran.

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	1,408 34th of 70	Single-core only
Navigation	1,355 32nd of 70	Single-core only
HTML5 Browser	1,249 36th of 70	Single-core only
PDF Viewer	1,407 36th of 70	Single-core only
Photo Library	1,240 34th of 70	Single-core only
Clang	1,101 46th of 70	Single-core only
Text Processing	1,163 37th of 70	Single-core only
Asset Compression	1,300 39th of 70	Single-core only
HDR	1,314 35th of 70	Single-core only
Photo Editor	1,501 35th of 70	Single-core only
Ray Tracer	1,209 40th of 70	Single-core only
Structure from Motion	1,555 31st of 70	Single-core only
Game Physics	1,183 40th of 70	Single-core only
Video Encoder	1,184 40th of 70	Single-core only
Audio Encoder	1,338 40th of 70	Single-core only
Video Player	1,376 41st 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	53 GB/s	3	23.3%
Scale higher is better	36.3 GB/s	3	14.7%
Add higher is better	39.1 GB/s	3	18.5%
Triad higher is better	39.9 GB/s	3	20.3%

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	410/s	3	1.53%
RSA 4096 verifies higher is better	27,279/s	3	0.389%
SHA-256 higher is better	1.5 GB/s	3	2.18%
AES-256-GCM higher is better	7.21 GB/s	3	1.17%
ChaCha20-Poly1305 higher is better	4.16 GB/s	3	0.956%

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	4.88 ms	3	0.683%
Requests per second, 90th percentile higher is better	400/s	3	0.289%
Request latency, 99.9th percentile lower is better	10.2 ms	3	5.63%
Requests per second, average higher is better	392/s	3	0.761%

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

Disk

Block size	Read	Write	Total
4K	105 MB/s (26,892 IOPS)	105.3 MB/s (26,963 IOPS)	210.4 MB/s (53,855 IOPS)
64K	1.09 GB/s (17,907 IOPS)	1.1 GB/s (18,002 IOPS)	2.19 GB/s (35,909 IOPS)
512K	1.4 GB/s (2,858 IOPS)	1.47 GB/s (3,010 IOPS)	2.87 GB/s (5,868 IOPS)
1M	1.65 GB/s (1,692 IOPS)	1.76 GB/s (1,805 IOPS)	3.42 GB/s (3,497 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	1.53 Gbps	2.96 Gbps
Eranium, Amsterdam, NL (100G)	IPv4	2.43 Gbps	2.71 Gbps
Leaseweb, Singapore, SG (10G)	IPv4	630 Mbps	895 Mbps
Clouvider, Los Angeles, CA, US (10G)	IPv4	2.53 Gbps	3.46 Gbps
Leaseweb, NYC, NY, US (10G)	IPv4	8.53 Gbps	7.64 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	3,775	0.265 ms
fdatasync	3,639	0.275 ms
fsync	2,116	0.473 ms
fsync_writethrough	Not supported here	n/a
open_sync	2,116	0.473 ms
Two 8 kB writes
open_datasync	1,872	0.534 ms
fdatasync	3,618	0.276 ms
fsync	2,032	0.492 ms
fsync_writethrough	Not supported here	n/a
open_sync	1,147	0.872 ms
open_sync at different write sizes
1 * 16kB open_sync write	2,109	0.474 ms
2 * 8kB open_sync writes	1,008	0.992 ms
4 * 4kB open_sync writes	549	1.82 ms
8 * 2kB open_sync writes	244	4.1 ms
16 * 1kB open_sync writes	127	7.87 ms
fsync on a non-write descriptor
write, fsync, close	1,963	0.509 ms
write, close, fsync	2,001	0.5 ms
Non-synced 8 kB writes
write	1,286,421	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	1,305	0.737 ms	0.942 ms	1.35 ms	3.55 ms
Read/write (TPC-B like)	8	4,395	1.64 ms	3.12 ms	4.48 ms	6.64 ms
Read/write (TPC-B like)	32	4,602	5.36 ms	16.7 ms	28 ms	46.4 ms
Read only	1	15,833	0.059 ms	0.093 ms	0.12 ms	0.206 ms
Read only	8	53,009	0.133 ms	0.25 ms	0.329 ms	0.645 ms
Read only	32	58,645	0.447 ms	0.929 ms	1.38 ms	2.41 ms

Scale 10 (1,000,000 accounts), loaded in 1.15 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	940	1.03 ms	1.42 ms	2.7 ms	4.94 ms
Read/write (TPC-B like)	8	3,568	2.05 ms	3.85 ms	5.75 ms	10.1 ms
Read/write (TPC-B like)	32	4,925	6.04 ms	10.6 ms	14.2 ms	21.1 ms
Read only	1	4,036	0.269 ms	0.468 ms	1.04 ms	2.24 ms
Read only	8	28,589	0.205 ms	0.658 ms	1.02 ms	2.41 ms
Read only	32	33,194	0.795 ms	1.81 ms	2.67 ms	4.43 ms

Scale 100 (10,000,000 accounts), loaded in 12.3 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	268	5,359	3.49 ms	5.09 ms	10.3 ms
Read/write	8	747	14,938	10.5 ms	16.4 ms	21.5 ms
Read/write	32	900	18,003	35.6 ms	56.8 ms	68.1 ms
Read only	1	677	10,831	1.42 ms	1.96 ms	2.43 ms
Read only	8	1,604	25,656	4.65 ms	7.57 ms	9.39 ms
Read only	32	1,502	24,038	20.4 ms	29.7 ms	35.6 ms

10 tables of 100,000 rows, loaded in 7.6 s. 60 s per run, one connection per thread.

Redis

memtier

Mix	Operations/s	Hits/s	Median	p99	p99.9
1:10 set:get	95,323	86,657	0.407 ms	2.64 ms	4.32 ms
1:1 set:get	88,158	44,079	0.471 ms	2.74 ms	4.38 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	357	1,490	1,963	1.87 s
Express express-postgresjs	7,005	10,101	25,277	88.7 ms
Fastify fastify-postgres	6,846	9,930	29,305	78.8 ms
NestJS nestjs	4,389	4,993	14,692	103 ms
Nuxt nuxt	6,194	9,736	28,567	145 ms
FastAPI fastapi	3,432	4,226	16,147	404 ms
Django django-postgresql	1,642	1,898	4,360	218 ms
Rails rails	2,815	3,817	13,391	118 ms
Laravel laravel	713	749	1,378	393 ms
Symfony symfony	966	1,060	2,792	298 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	789 ms	928 ms	961 ms	0%
Shop page	2	1.09 s	1.26 s	1.3 s	0%
Category page	2	657 ms	1.1 s	1.17 s	0%
Product page	2	746 ms	893 ms	946 ms	0%
Product search	2	642 ms	772 ms	801 ms	0%
32 virtual users
Home page	1.98	2.92 s	3.33 s	3.43 s	0%
Shop page	1.98	3.89 s	4.49 s	4.65 s	0%
Category page	1.98	3.41 s	4.14 s	4.27 s	0%
Product page	1.98	2.99 s	3.25 s	3.32 s	0%
Product search	1.98	2.72 s	2.99 s	3.14 s	0%

Cart and checkout

Step	Requests/s	Median	p95	p99	Error rate
8 virtual users
Add to cart	3.81	875 ms	1.04 s	1.13 s	0%
Cart	3.81	700 ms	815 ms	875 ms	0%
Checkout	3.81	534 ms	636 ms	668 ms	0%
32 virtual users
Add to cart	3.95	2.78 s	3.4 s	3.52 s	0%
Cart	3.95	2.89 s	3.17 s	3.27 s	0%
Checkout	3.95	2.32 s	2.68 s	2.78 s	0%

k6 with a fixed number of virtual users, 61 to 65 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	2 GB	1.9 GiB
Storage	40 GB NVMe	37.5 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-Rome Processor
CPU cores: 2
CPU frequency: 2445 MHz
AES-NI: Yes
VM-x/AMD-V: No
RAM reported: 1.9 GiB
Swap reported: 0 MiB
Disk reported: 37.5 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: CPX11
Product line: regular_purpose
Location: Ashburn, VA
vCPU: 2 shared vCPU
Architecture: x86_64
RAM: 2 GB
Storage: 40 GB NVMe
Virtualization: KVM
Network: Unknown
Transfer: 1.1 TB
Port 25: On request
Monthly price at test time: $20.49
Hourly price at test time: $0.033
YABS version: v2026-09-20
Ran at: Oct 10, 2026, 17:02:33 UTC
Parser version: 5

Run timeline

Queued: Oct 10, 2026, 17:02:00 UTC
First update from the server: Oct 10, 2026, 17:02:28 UTC
Completed: Oct 10, 2026, 18:12:13 UTC
Server deleted: Oct 10, 2026, 18:12:13 UTC
Cost of this run: $0.038

TestResultTimevs last run

Watch party 0 watching

This party closed at 18:09 UTC. It's read only now.

Loading the party

RackCensus

Server

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