Skip to main content

UpCloud Premium 2 vCPU / 2 GB

Frankfurt #1

Benchmark complete
Finished Oct 17, 2026
Geekbench 7 single-core
2,333
Faster than 97% of 2 vCPU plans
Geekbench 7 multi-core
Skipped
Disk 4K IOPS
31,344
fio random read and write, 50/50
Monthly price at test time
€18.00 EUR
Same price today

Server

  • Provider
    UpCloud
  • Plan
    Premium 2 vCPU / 2 GB
  • Location
    Frankfurt #1
  • vCPU
    2
  • RAM
    2 GB
  • Storage
    50 GB NVMe
  • Network
    Unknown
  • OS image
    01000000-0000-4000-8000-000030240200
  • Trigger
    Scheduled
  • Queued at
    Oct 17, 2026, 11:56:00 UTC
Visit UpCloud

Referral link. If you sign up through it, I may get paid. How this works

Latest runs of this plan

Full history of this plan

CPU

Version Single-core Multi-core
Geekbench 7 2,333 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
2,433
2nd of 241
Single-core only
Navigation
1,534
59th of 241
Single-core only
HTML5 Browser
2,280
6th of 241
Single-core only
PDF Viewer
2,758
2nd of 241
Single-core only
Photo Library
2,665
1st of 241
Single-core only
Clang
2,216
3rd of 241
Single-core only
Text Processing
2,584
1st of 241
Single-core only
Asset Compression
1,981
13th of 241
Single-core only
HDR
2,368
6th of 241
Single-core only
Photo Editor
3,539
2nd of 241
Single-core only
Ray Tracer
1,887
9th of 241
Single-core only
Structure from Motion
2,761
2nd of 241
Single-core only
Game Physics
1,508
47th of 241
Single-core only
Video Encoder
2,451
3rd of 241
Single-core only
Audio Encoder
2,502
2nd of 241
Single-core only
Video Player
2,693
6th of 241
Single-core only

Rank among the 241 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 75 GB/s 3 17.6%
Scale higher is better 59.3 GB/s 3 23.9%
Add higher is better 63 GB/s 3 14.1%
Triad higher is better 66.7 GB/s 3 19.8%

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 1,806/s 3 1.29%
RSA 4096 verifies higher is better 49,392/s 3 0.185%
SHA-256 higher is better 2.57 GB/s 3 16.3%
AES-256-GCM higher is better 46.7 GB/s 3 0.117%
ChaCha20-Poly1305 higher is better 11.1 GB/s 3 13.1%

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 2.66 ms 3 1.66%
Requests per second, 90th percentile higher is better 749/s 3 0.963%
Request latency, 99.9th percentile lower is better 5.24 ms 3 25%
Requests per second, average higher is better 702/s 3 8.13%

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

Disk

Block size Read Write Total
4K 61.2 MB/s (15,657 IOPS) 61.3 MB/s (15,687 IOPS) 122.4 MB/s (31,344 IOPS)
64K 399.8 MB/s (6,396 IOPS) 401.9 MB/s (6,430 IOPS) 801.7 MB/s (12,826 IOPS)
512K 382 MB/s (764 IOPS) 402.3 MB/s (804 IOPS) 784.4 MB/s (1,568 IOPS)
1M 377.4 MB/s (377 IOPS) 402.5 MB/s (402 IOPS) 779.9 MB/s (779 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 1.06 Gbps 1.06 Gbps
Eranium, Amsterdam, NL (100G) IPv4 1.07 Gbps 1.07 Gbps
Leaseweb, Singapore, SG (10G) IPv4 934 Mbps 930 Mbps
Clouvider, Los Angeles, CA, US (10G) IPv4 936 Mbps 139 Mbps
Leaseweb, NYC, NY, US (10G) IPv4 1.02 Gbps 980 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 9,042 0.111 ms
fdatasync 6,750 0.148 ms
fsync 4,341 0.23 ms
fsync_writethrough Not supported here n/a
open_sync 8,377 0.119 ms
Two 8 kB writes
open_datasync 4,831 0.207 ms
fdatasync 9,383 0.107 ms
fsync 7,690 0.13 ms
fsync_writethrough Not supported here n/a
open_sync 4,343 0.23 ms
open_sync at different write sizes
1 * 16kB open_sync write 6,792 0.147 ms
2 * 8kB open_sync writes 4,132 0.242 ms
4 * 4kB open_sync writes 2,117 0.472 ms
8 * 2kB open_sync writes 1,053 0.95 ms
16 * 1kB open_sync writes 486 2.06 ms
fsync on a non-write descriptor
write, fsync, close 5,398 0.185 ms
write, close, fsync 4,965 0.201 ms
Non-synced 8 kB writes
write 2,797,764 0 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 4,302 0.181 ms 0.475 ms 0.673 ms 2.91 ms 0
Read/write (TPC-B like) 8 11,896 0.513 ms 1.34 ms 2.29 ms 14.7 ms 0
Read/write (TPC-B like) 32 12,400 1.76 ms 6.69 ms 13.9 ms 36.6 ms 0
Read only 1 56,671 0.018 ms 0.021 ms 0.026 ms 0.047 ms 0
Read only 8 157,926 0.044 ms 0.082 ms 0.112 ms 0.166 ms 0
Read only 32 131,663 0.206 ms 0.341 ms 0.427 ms 0.671 ms 0

Scale 10 (1,000,000 accounts), loaded in 0.9 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.

pgbench at scale 100

Workload Clients Transactions/s Median p95 p99 p99.9 Failed
Read/write (TPC-B like) 1 2,399 0.318 ms 0.779 ms 1.29 ms 4.59 ms 0
Read/write (TPC-B like) 8 7,127 0.724 ms 2.14 ms 4.98 ms 66.8 ms 0
Read/write (TPC-B like) 32 10,084 2.32 ms 4.96 ms 27.2 ms 78.7 ms 0
Read only 1 23,280 0.023 ms 0.092 ms 0.167 ms 0.666 ms 0
Read only 8 90,196 0.059 ms 0.175 ms 0.267 ms 2.06 ms 0
Read only 32 87,948 0.252 ms 0.56 ms 1.17 ms 24 ms 0

Scale 100 (10,000,000 accounts), loaded in 14 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 1 847 16,935 0.971 ms 1.73 ms 3.49 ms 0
Read/write 8 1,295 25,890 5.28 ms 10.5 ms 30.8 ms 0
Read/write 32 2,422 48,441 11.4 ms 27.2 ms 54.8 ms 0
Read only 1 2,833 45,332 0.291 ms 0.566 ms 0.782 ms 0
Read only 8 5,669 90,703 1.27 ms 2.31 ms 3.13 ms 0
Read only 32 4,465 71,436 6.67 ms 10.7 ms 12.8 ms 0

10 tables of 100,000 rows, loaded in 6.4 s. 60 s per run, one connection per thread.

Redis

memtier

Mix Operations/s Hits/s Median p99 p99.9
1:10 set:get 155,017 140,924 0.231 ms 2.3 ms 3.78 ms
1:1 set:get 177,693 88,846 0.207 ms 2.26 ms 3.22 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 885 3,700 4,206 511 ms
Express express-postgresjs 11,815 19,416 136,874 114 ms
Fastify fastify-postgres 10,748 22,324 143,540 42.1 ms
NestJS nestjs 13,194 14,837 91,721 58 ms
Nuxt nuxt 10,824 20,529 86,346 84.1 ms
FastAPI fastapi 10,213 12,594 50,151 103 ms
Django django-postgresql 3,688 4,733 13,859 99.7 ms
Rails rails 6,623 10,121 49,363 76.9 ms
Laravel laravel 1,794 1,748 3,163 320 ms
Symfony symfony 2,017 2,255 7,815 139 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.

Web apps

WooCommerce

Page mix

Step Requests/s Median p95 p99 Error rate
8 virtual users
Home page 3.97 382 ms 562 ms 629 ms 0%
Shop page 3.97 526 ms 778 ms 869 ms 0%
Category page 3.97 332 ms 561 ms 688 ms 0%
Product page 3.97 359 ms 513 ms 602 ms 0%
Product search 3.97 316 ms 463 ms 583 ms 0%
32 virtual users
Home page 4.03 1.4 s 1.72 s 1.84 s 0%
Shop page 4.03 1.85 s 2.15 s 2.32 s 0%
Category page 4.03 1.6 s 2.14 s 2.41 s 0%
Product page 4.03 1.4 s 2.39 s 2.52 s 0%
Product search 4.03 1.31 s 1.99 s 2.16 s 0%

Cart and checkout

Step Requests/s Median p95 p99 Error rate
8 virtual users
Add to cart 7.85 417 ms 576 ms 705 ms 0%
Cart 7.85 340 ms 458 ms 638 ms 0%
Checkout 7.85 262 ms 344 ms 473 ms 0%
32 virtual users
Add to cart 7.91 1.35 s 1.98 s 2.34 s 0%
Cart 7.91 1.39 s 1.99 s 2.16 s 0%
Checkout 7.91 1.12 s 1.53 s 1.71 s 0%

k6 with a fixed number of virtual users, 61 to 64 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 4.13 98.6 ms 183 ms 311 ms 0%
Topic JSON 2.08 50.3 ms 95.3 ms 120 ms 0%
Home page 2.08 87.6 ms 174 ms 304 ms 0%
Categories page 1.03 103 ms 177 ms 196 ms 0%
User activity page 1.03 62.8 ms 121 ms 158 ms 0%
4 virtual users
Topic page 7.37 213 ms 380 ms 536 ms 0%
Topic JSON 3.69 157 ms 276 ms 398 ms 0%
Home page 3.68 208 ms 357 ms 404 ms 0%
Categories page 1.86 217 ms 362 ms 454 ms 0%
User activity page 1.86 171 ms 296 ms 362 ms 0%
16 virtual users
Topic page 7.6 772 ms 1.24 s 1.36 s 0%
Topic JSON 3.86 722 ms 1.15 s 1.27 s 0%
Home page 3.81 768 ms 1.19 s 1.31 s 0%
Categories page 1.94 773 ms 1.25 s 1.32 s 0%
User activity page 1.89 733 ms 1.21 s 1.27 s 0%

Anonymous

Step Requests/s Median p95 p99 Error rate
16 virtual users
Topic page 171 33.1 ms 71.6 ms 100 ms 0%
Topic JSON 85.6 33.6 ms 67.6 ms 96.3 ms 0%
Home page 85.6 34.1 ms 73.2 ms 104 ms 0%
Categories page 42.8 33.9 ms 67.9 ms 94.5 ms 0%
User activity page 42.8 60.5 ms 114 ms 149 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 2 app workers and 128 MB of Postgres shared buffers, what its installer picks for a 1 GB box. It added a 2 GB swapfile first, the way its installer does, and used up to 302 MB of it. The container peaked at 1,320 MB.

Ghost

Readers

Step Requests/s Median p95 p99 Error rate
1 virtual user
Home page 16.8 14.5 ms 26.9 ms 31.1 ms 0%
Post page 41.9 8.69 ms 15.4 ms 20 ms 0%
Tag page 8.38 10.5 ms 18.4 ms 24.6 ms 0%
Author page 8.38 10.5 ms 18.8 ms 23.3 ms 0%
RSS feed 8.38 12 ms 21.2 ms 24.8 ms 0%
4 virtual users
Home page 20.6 46.1 ms 82.1 ms 109 ms 0%
Post page 51.4 35 ms 61.9 ms 83.3 ms 0%
Tag page 10.3 30 ms 52 ms 69.8 ms 0%
Author page 10.3 29.6 ms 50.5 ms 69.8 ms 0%
RSS feed 10.3 32.1 ms 60.6 ms 76.9 ms 0%
16 virtual users
Home page 20.3 184 ms 322 ms 383 ms 0%
Post page 50.8 153 ms 265 ms 324 ms 0%
Tag page 10.2 101 ms 171 ms 219 ms 0%
Author page 10.2 98.8 ms 178 ms 230 ms 0%
RSS feed 10.1 101 ms 178 ms 228 ms 0%

Staff in Ghost Admin

Step Requests/s Median p95 p99 Error rate
1 virtual user
Post list 9.4 14.5 ms 25.9 ms 36 ms 0%
Post editor 28.2 6.13 ms 10.3 ms 14.2 ms 0%
Saving a draft 28.2 14.9 ms 25 ms 32.4 ms 0%
Member list 9.4 16.9 ms 29.5 ms 33.2 ms 0%
4 virtual users
Post list 11 20.4 ms 35.6 ms 50.4 ms 0%
Post editor 33.1 10.3 ms 19 ms 31.6 ms 0%
Saving a draft 33.2 84.9 ms 144 ms 179 ms 0%
Member list 11 22 ms 38.6 ms 50.5 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 387 MB and MySQL at 586 MB.

Wagtail

Visitors

Step Requests/s Median p95 p99 Error rate
1 virtual user
Home page 3.43 36.2 ms 63.3 ms 69.4 ms 0%
Blog index 3.43 110 ms 200 ms 217 ms 0%
Blog post 10.3 12.4 ms 21.5 ms 27.6 ms 0%
Bread page 10.3 13.3 ms 23.7 ms 26.2 ms 0%
Location page 3.43 12.7 ms 23.5 ms 25.6 ms 0%
Site search 3.43 22.6 ms 42.4 ms 55.5 ms 0%
4 virtual users
Home page 5.31 101 ms 179 ms 232 ms 0%
Blog index 5.34 330 ms 534 ms 641 ms 0%
Blog post 16 30.8 ms 57.1 ms 81.2 ms 0%
Bread page 16 33.5 ms 58.2 ms 78.9 ms 0%
Location page 5.34 31.2 ms 52.7 ms 72.7 ms 0%
Site search 5.31 50.5 ms 115 ms 178 ms 0%
16 virtual users
Home page 6.02 300 ms 471 ms 564 ms 0%
Blog index 6.1 700 ms 1.07 s 1.26 s 0%
Blog post 18.1 171 ms 332 ms 431 ms 0%
Bread page 18.2 176 ms 316 ms 404 ms 0%
Location page 6.08 173 ms 303 ms 404 ms 0%
Site search 6.03 214 ms 369 ms 453 ms 0%

Editors in the Wagtail admin

Step Requests/s Median p95 p99 Error rate
1 virtual user
Admin dashboard 1.51 39.7 ms 67.4 ms 75.4 ms 0%
Page explorer 3.06 139 ms 267 ms 319 ms 0%
Page editor 4.55 69.2 ms 126 ms 149 ms 0%
Image library 1.51 28.3 ms 50.1 ms 94.8 ms 0%
Admin page search 1.51 20.3 ms 37.5 ms 184 ms 0%
4 virtual users
Admin dashboard 2.3 128 ms 315 ms 405 ms 0%
Page explorer 4.64 366 ms 706 ms 806 ms 0%
Page editor 6.96 185 ms 398 ms 523 ms 0%
Image library 2.3 106 ms 301 ms 352 ms 0%
Admin page search 2.3 50 ms 161 ms 201 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. 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 310 MB, Postgres at 178 MB, and Redis at 26 MB.

Advertised vs. measured

Advertised Measured
vCPU 2 2
CPU Shared vCPU 0.1% of busy CPU time stolen
RAM 2 GB 1.8 GiB
Storage 50 GB NVMe 49.1 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 9575F 64-Core Processor
CPU cores
2
CPU frequency
3295 MHz
AES-NI
Yes
VM-x/AMD-V
No
RAM reported
1.8 GiB
Swap reported
0 MiB
Disk reported
49.1 GiB
Virtualization reported
KVM
CPU steal
0.1% of busy CPU time stolen
Distro
Ubuntu 24.04.4 LTS
Kernel
6.8.0-137-generic
Architecture
x86_64
IPv4 reachable
Yes
IPv6 reachable
Yes

Advertised at test time

Plan
Premium 2 vCPU / 2 GB
Product line
Premium
Location
Frankfurt #1
vCPU
2 shared vCPU
Architecture
x86_64
RAM
2 GB
Storage
50 GB NVMe
Virtualization
KVM
Network
Unknown
Transfer
2.05 TB
Port 25
Unknown
Monthly price at test time
€18.00 EUR
Hourly price at test time
€0.027 EUR
YABS version
v2026-09-20
Ran at
Oct 17, 2026, 11:59:51 UTC
Parser version
5

Run timeline

Queued
Oct 17, 2026, 11:56:00 UTC
First update from the server
Oct 17, 2026, 11:59:45 UTC
Completed
Oct 17, 2026, 13:25:30 UTC
Server deleted
Oct 17, 2026, 13:25:30 UTC
Cost of this run
€0.040 EUR
Watch party 0 watching

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

Watch party

Loading the party