Skip to main content

UpCloud Premium 4 vCPU / 8 GB

San Jose #1

Benchmark complete
Finished Oct 14, 2026
Geekbench 7 single-core
2,125
Faster than 100% of 4 vCPU plans
Geekbench 7 multi-core
7,645
Faster than 100% of 4 vCPU plans
Disk 4K IOPS
32,881
fio random read and write, 50/50
Monthly price at test time
€54.00 EUR
121 multi-core points per dollar
Same price today

Server

  • Provider
    UpCloud
  • Plan
    Premium 4 vCPU / 8 GB
  • Location
    San Jose #1
  • vCPU
    4
  • RAM
    8 GB
  • Storage
    100 GB NVMe
  • Network
    Unknown
  • OS image
    01000000-0000-4000-8000-000030240200
  • Trigger
    Scheduled
  • Queued at
    Oct 14, 2026, 08:24:00 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 2,125 7,645 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
2,117
4th of 170
6,318
5th of 124
Navigation
1,626
32nd of 170
Single-core only
HTML5 Browser
2,149
4th of 170
Single-core only
PDF Viewer
2,420
3rd of 170
Single-core only
Photo Library
2,104
4th of 170
8,394
8th of 124
Clang
1,769
5th of 170
6,692
15th of 124
Text Processing
1,935
5th of 170
7,221
10th of 124
Asset Compression
2,214
3rd of 170
7,957
14th of 124
HDR
2,304
3rd of 170
7,254
5th of 124
Photo Editor
3,337
2nd of 170
11,411
1st of 124
Ray Tracer
1,596
29th of 170
6,915
16th of 124
Structure from Motion
2,344
4th of 170
Single-core only
Game Physics
1,894
5th of 170
Single-core only
Video Encoder
2,033
6th of 170
Single-core only
Audio Encoder
2,227
3rd of 170
Single-core only
Video Player
2,453
4th of 170
Single-core only

Rank among the 170 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 114 GB/s 3 12.6%
Scale higher is better 91.6 GB/s 3 26.8%
Add higher is better 96.8 GB/s 3 27.8%
Triad higher is better 94.9 GB/s 3 24.8%

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 2,935/s 3 13.9%
RSA 4096 verifies higher is better 77,309/s 3 11.8%
SHA-256 higher is better 4.61 GB/s 3 3.91%
AES-256-GCM higher is better 76.3 GB/s 3 0.902%
ChaCha20-Poly1305 higher is better 22.1 GB/s 3 1.11%

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 3.26 ms 3 0.993%
Requests per second, 90th percentile higher is better 1,150/s 3 1.84%
Request latency, 99.9th percentile lower is better 6.47 ms 3 1.89%
Requests per second, average higher is better 1,084/s 3 0.953%

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

Disk

Block size Read Write Total
4K 64.2 MB/s (16,423 IOPS) 64.3 MB/s (16,458 IOPS) 128.4 MB/s (32,881 IOPS)
64K 373.3 MB/s (5,972 IOPS) 375.3 MB/s (6,004 IOPS) 748.5 MB/s (11,976 IOPS)
512K 374.3 MB/s (748 IOPS) 394.2 MB/s (788 IOPS) 768.5 MB/s (1,536 IOPS)
1M 379.3 MB/s (379 IOPS) 404.5 MB/s (404 IOPS) 783.8 MB/s (783 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 954 Mbps 985 Mbps
Eranium, Amsterdam, NL (100G) IPv4 965 Mbps 970 Mbps
Leaseweb, Singapore, SG (10G) IPv4 942 Mbps 859 Mbps
Clouvider, Los Angeles, CA, US (10G) IPv4 1.06 Gbps 1.07 Gbps
Leaseweb, NYC, NY, US (10G) IPv4 1.03 Gbps 1.02 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 6,429 0.156 ms
fdatasync 5,628 0.178 ms
fsync 3,047 0.328 ms
fsync_writethrough Not supported here n/a
open_sync 3,777 0.265 ms
Two 8 kB writes
open_datasync 2,914 0.343 ms
fdatasync 4,768 0.21 ms
fsync 2,838 0.352 ms
fsync_writethrough Not supported here n/a
open_sync 1,790 0.559 ms
open_sync at different write sizes
1 * 16kB open_sync write 2,881 0.347 ms
2 * 8kB open_sync writes 1,862 0.537 ms
4 * 4kB open_sync writes 901 1.11 ms
8 * 2kB open_sync writes 428 2.34 ms
16 * 1kB open_sync writes 230 4.35 ms
fsync on a non-write descriptor
write, fsync, close 3,499 0.286 ms
write, close, fsync 3,353 0.298 ms
Non-synced 8 kB writes
write 2,760,224 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 3,109 0.275 ms 0.483 ms 0.927 ms 4.28 ms 0
Read/write (TPC-B like) 8 10,370 0.591 ms 1.5 ms 2.78 ms 13.1 ms 0
Read/write (TPC-B like) 32 12,611 1.61 ms 7.36 ms 15.1 ms 41.9 ms 0
Read only 1 69,983 0.013 ms 0.021 ms 0.026 ms 0.056 ms 0
Read only 8 248,274 0.028 ms 0.061 ms 0.079 ms 0.112 ms 0
Read only 32 238,712 0.119 ms 0.209 ms 0.261 ms 0.358 ms 0

Scale 10 (1,000,000 accounts), loaded in 0.81 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 2,409 0.35 ms 0.577 ms 1.46 ms 10.8 ms 0
Read/write (TPC-B like) 8 8,923 0.547 ms 1.39 ms 5.32 ms 70.2 ms 0
Read/write (TPC-B like) 32 12,401 1.58 ms 4.43 ms 20 ms 85.6 ms 0
Read only 1 53,501 0.016 ms 0.027 ms 0.035 ms 0.07 ms 0
Read only 8 233,196 0.032 ms 0.06 ms 0.081 ms 0.127 ms 0
Read only 32 195,766 0.147 ms 0.262 ms 0.338 ms 0.493 ms 0

Scale 100 (10,000,000 accounts), loaded in 10.4 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 1 434 8,678 1.96 ms 3.3 ms 8.43 ms 0
Read/write 8 1,656 33,122 3.89 ms 8.43 ms 34.3 ms 0
Read/write 32 2,835 56,706 9.91 ms 19.6 ms 55.8 ms 0
Read only 1 2,308 36,933 0.409 ms 0.619 ms 0.811 ms 0
Read only 8 7,796 124,735 0.989 ms 1.58 ms 2.22 ms 0
Read only 32 7,312 116,985 4.25 ms 6.1 ms 7.43 ms 0

10 tables of 100,000 rows, loaded in 4.6 s. 60 s per run, one connection per thread.

Redis

memtier

Mix Operations/s Hits/s Median p99 p99.9
1:10 set:get 204,903 186,275 0.311 ms 0.599 ms 0.959 ms
1:1 set:get 204,611 102,305 0.311 ms 0.559 ms 0.799 ms

1,000,000 keys of 32 bytes. 64 connections over 4 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 1,598 6,205 9,716 304 ms
Express express-postgresjs 19,569 30,569 197,751 25.5 ms
Fastify fastify-postgres 18,561 31,811 223,421 30.2 ms
NestJS nestjs 20,818 23,412 128,469 30.7 ms
Nuxt nuxt 16,485 32,755 151,740 42.5 ms
FastAPI fastapi 13,755 21,021 72,944 61.6 ms
Django django-postgresql 6,050 6,888 16,592 99 ms
Rails rails 9,065 14,718 75,772 56.6 ms
Laravel laravel 2,341 2,491 4,212 147 ms
Symfony symfony 3,011 3,129 10,415 96.9 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 6.33 250 ms 345 ms 395 ms 0%
Shop page 6.33 329 ms 460 ms 517 ms 0%
Category page 6.33 216 ms 363 ms 408 ms 0%
Product page 6.33 241 ms 344 ms 387 ms 0%
Product search 6.33 204 ms 300 ms 348 ms 0%
32 virtual users
Home page 5.69 1.08 s 1.34 s 1.4 s 0%
Shop page 5.69 1.36 s 1.59 s 1.69 s 0%
Category page 5.69 1.08 s 1.41 s 1.56 s 0%
Product page 5.69 1.05 s 1.27 s 1.37 s 0%
Product search 5.69 964 ms 1.19 s 1.29 s 0%

Cart and checkout

Step Requests/s Median p95 p99 Error rate
8 virtual users
Add to cart 12.4 255 ms 349 ms 414 ms 0%
Cart 12.4 218 ms 296 ms 341 ms 0%
Checkout 12.4 172 ms 236 ms 281 ms 0%
32 virtual users
Add to cart 12 989 ms 1.2 s 1.32 s 0%
Cart 12 908 ms 1.12 s 1.23 s 0%
Checkout 12 752 ms 900 ms 997 ms 0%

k6 with a fixed number of virtual users, 61 to 62 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 2.99 146 ms 181 ms 427 ms 0%
Topic JSON 1.51 73.9 ms 87.6 ms 95.4 ms 0%
Home page 1.51 129 ms 252 ms 431 ms 0%
Categories page 0.747 150 ms 169 ms 371 ms 0%
User activity page 0.747 90.4 ms 224 ms 285 ms 0%
4 virtual users
Topic page 10.6 162 ms 261 ms 438 ms 0%
Topic JSON 5.28 81 ms 152 ms 260 ms 0%
Home page 5.28 145 ms 247 ms 411 ms 0%
Categories page 2.65 169 ms 250 ms 510 ms 0%
User activity page 2.65 104 ms 194 ms 347 ms 0%
16 virtual users
Topic page 9.34 703 ms 1.02 s 1.47 s 0%
Topic JSON 4.69 522 ms 747 ms 867 ms 0%
Home page 4.72 645 ms 899 ms 1.25 s 0%
Categories page 2.36 721 ms 1.26 s 1.42 s 0%
User activity page 2.36 569 ms 745 ms 853 ms 0%

Anonymous

Step Requests/s Median p95 p99 Error rate
16 virtual users
Topic page 196 19.6 ms 42 ms 63.9 ms 0%
Topic JSON 97.8 18.6 ms 40.3 ms 59.9 ms 0%
Home page 97.8 19.8 ms 40.9 ms 61.9 ms 0%
Categories page 48.9 19.6 ms 41 ms 58.9 ms 0%
User activity page 48.9 124 ms 174 ms 290 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 8 app workers and 2048 MB of Postgres shared buffers, what its installer picks for a 8 GB box. Its installer adds no swap on a box this big. The container peaked at 2,837 MB.

Ghost

Readers

Step Requests/s Median p95 p99 Error rate
1 virtual user
Home page 12.7 22.7 ms 27.5 ms 32.7 ms 0%
Post page 31.7 13.2 ms 16.3 ms 18.7 ms 0%
Tag page 6.35 13.7 ms 17.3 ms 19.6 ms 0%
Author page 6.33 14.7 ms 17.9 ms 20 ms 0%
RSS feed 6.33 14.1 ms 18 ms 22.2 ms 0%
4 virtual users
Home page 17.3 61.3 ms 78.5 ms 102 ms 0%
Post page 43.2 42.7 ms 60.3 ms 75.5 ms 0%
Tag page 8.64 37.7 ms 50.7 ms 61.9 ms 0%
Author page 8.64 39.2 ms 53.3 ms 63 ms 0%
RSS feed 8.63 39.8 ms 53.9 ms 71.6 ms 0%
16 virtual users
Home page 18.3 224 ms 266 ms 304 ms 0%
Post page 45.6 183 ms 224 ms 266 ms 0%
Tag page 9.17 123 ms 156 ms 178 ms 0%
Author page 9.12 126 ms 156 ms 181 ms 0%
RSS feed 9.11 126 ms 160 ms 179 ms 0%

Staff in Ghost Admin

Step Requests/s Median p95 p99 Error rate
1 virtual user
Post list 7.25 18.4 ms 23 ms 25 ms 0%
Post editor 21.8 8.8 ms 11.9 ms 14.3 ms 0%
Saving a draft 21.8 21.7 ms 28.8 ms 39.3 ms 0%
Member list 7.25 21 ms 25.9 ms 32.7 ms 0%
4 virtual users
Post list 10.1 23.4 ms 29.4 ms 37.3 ms 0%
Post editor 30.4 12.6 ms 17.1 ms 20.9 ms 0%
Saving a draft 30.4 99.4 ms 125 ms 150 ms 0%
Member list 10.1 25.6 ms 31.8 ms 37.1 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 547 MB and MySQL at 626 MB.

Wagtail

Visitors

Step Requests/s Median p95 p99 Error rate
1 virtual user
Home page 2.53 53.9 ms 68 ms 88.7 ms 0%
Blog index 2.53 162 ms 191 ms 279 ms 0%
Blog post 7.57 18.2 ms 27 ms 30.5 ms 0%
Bread page 7.59 19.6 ms 24 ms 26.9 ms 0%
Location page 2.53 18.8 ms 23 ms 25.4 ms 0%
Site search 2.53 30.2 ms 50.9 ms 55.6 ms 0%
4 virtual users
Home page 4.04 136 ms 222 ms 344 ms 0%
Blog index 4.07 546 ms 709 ms 1.11 s 0%
Blog post 12.2 36.4 ms 68.5 ms 96.4 ms 0%
Bread page 12.2 39.6 ms 65.8 ms 86.5 ms 0%
Location page 4.09 38 ms 58.5 ms 74.6 ms 0%
Site search 4.06 57.7 ms 134 ms 208 ms 0%
16 virtual users
Home page 5.05 366 ms 538 ms 715 ms 0%
Blog index 5.1 951 ms 1.26 s 1.4 s 0%
Blog post 15.1 201 ms 376 ms 507 ms 0%
Bread page 15.2 203 ms 361 ms 509 ms 0%
Location page 5.1 192 ms 344 ms 441 ms 0%
Site search 5.01 236 ms 398 ms 478 ms 0%

Editors in the Wagtail admin

Step Requests/s Median p95 p99 Error rate
1 virtual user
Admin dashboard 1.14 60.2 ms 67.1 ms 85.5 ms 0%
Page explorer 2.32 188 ms 234 ms 374 ms 0%
Page editor 3.48 98.5 ms 138 ms 309 ms 0%
Image library 1.16 38.8 ms 46.4 ms 322 ms 0%
Admin page search 1.16 26.2 ms 53.7 ms 63.7 ms 0%
4 virtual users
Admin dashboard 1.87 172 ms 390 ms 492 ms 0%
Page explorer 3.75 480 ms 799 ms 922 ms 0%
Page editor 5.67 238 ms 471 ms 541 ms 0%
Image library 1.9 174 ms 310 ms 496 ms 0%
Admin page search 1.87 61.5 ms 243 ms 365 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 305 MB, Postgres at 151 MB, and Redis at 26 MB.

Advertised vs. measured

Advertised Measured
vCPU 4 4
CPU Shared vCPU 0.2% of busy CPU time stolen
RAM 8 GB 7.7 GiB
Storage 100 GB NVMe 98.3 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
4
CPU frequency
3295 MHz
AES-NI
Yes
VM-x/AMD-V
No
RAM reported
7.7 GiB
Swap reported
0 MiB
Disk reported
98.3 GiB
Virtualization reported
KVM
CPU steal
0.2% 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 4 vCPU / 8 GB
Product line
Premium
Location
San Jose #1
vCPU
4 shared vCPU
Architecture
x86_64
RAM
8 GB
Storage
100 GB NVMe
Virtualization
KVM
Network
Unknown
Transfer
5.12 TB
Port 25
Unknown
Monthly price at test time
€54.00 EUR
Hourly price at test time
€0.080 EUR
YABS version
v2026-09-20
Ran at
Oct 14, 2026, 08:25:53 UTC
Parser version
5

Run timeline

Queued
Oct 14, 2026, 08:24:00 UTC
Server created
Oct 14, 2026, 08:25:48 UTC
First update from the server
Oct 14, 2026, 08:25:48 UTC
Completed
Oct 14, 2026, 09:54:05 UTC
Server deleted
Oct 14, 2026, 09:54:05 UTC
Cost of this run
€0.12 EUR
Watch party 0 watching

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

Watch party

Loading the party