Skip to main content

UpCloud Premium 2 vCPU / 2 GB

New York #1

Benchmark complete
Finished Oct 10, 2026
Geekbench 7 single-core
1,886
Faster than 100% of 2 vCPU plans
Geekbench 7 multi-core
Skipped
Disk 4K IOPS
27,227
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
    New York #1
  • vCPU
    2
  • RAM
    2 GB
  • Storage
    50 GB NVMe
  • Network
    Unknown
  • OS image
    01000000-0000-4000-8000-000030240200
  • Trigger
    Scheduled
  • Queued at
    Oct 9, 2026, 23:20:00 UTC
Visit UpCloud

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

Skip the referral How this works

CPU

Version Single-core Multi-core
Geekbench 7 1,886 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
1,911
2nd of 61
Single-core only
Navigation
1,339
29th of 61
Single-core only
HTML5 Browser
1,751
12th of 61
Single-core only
PDF Viewer
2,097
4th of 61
Single-core only
Photo Library
2,103
2nd of 61
Single-core only
Clang
1,658
8th of 61
Single-core only
Text Processing
1,701
9th of 61
Single-core only
Asset Compression
1,967
2nd of 61
Single-core only
HDR
2,042
2nd of 61
Single-core only
Photo Editor
2,830
2nd of 61
Single-core only
Ray Tracer
1,308
33rd of 61
Single-core only
Structure from Motion
2,044
2nd of 61
Single-core only
Game Physics
1,700
11th of 61
Single-core only
Video Encoder
1,789
10th of 61
Single-core only
Audio Encoder
2,064
2nd of 61
Single-core only
Video Player
2,401
2nd of 61
Single-core only

Rank among the 61 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 70.1 GB/s 3 9.97%
Scale higher is better 60 GB/s 3 2.99%
Add higher is better 62.4 GB/s 3 3.1%
Triad higher is better 61.5 GB/s 3 7.17%

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 1,532/s 3 2.99%
RSA 4096 verifies higher is better 41,524/s 3 2.43%
SHA-256 higher is better 2.23 GB/s 3 5.64%
AES-256-GCM higher is better 35 GB/s 3 4.77%
ChaCha20-Poly1305 higher is better 10.7 GB/s 3 1.4%

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 3.46 ms 3 2.86%
Requests per second, 90th percentile higher is better 590/s 3 1.6%
Request latency, 99.9th percentile lower is better 6.7 ms 3 11.1%
Requests per second, average higher is better 538/s 3 5.26%

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

Disk

Block size Read Write Total
4K 53.1 MB/s (13,600 IOPS) 53.2 MB/s (13,627 IOPS) 106.4 MB/s (27,227 IOPS)
64K 363.5 MB/s (5,815 IOPS) 365.4 MB/s (5,845 IOPS) 728.8 MB/s (11,660 IOPS)
512K 382.3 MB/s (764 IOPS) 402.6 MB/s (805 IOPS) 785 MB/s (1,569 IOPS)
1M 378.8 MB/s (378 IOPS) 404 MB/s (403 IOPS) 782.7 MB/s (781 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 956 Mbps 1 Gbps
Eranium, Amsterdam, NL (100G) IPv4 959 Mbps 1.02 Gbps
Leaseweb, Singapore, SG (10G) IPv4 495 Mbps 901 Mbps
Clouvider, Los Angeles, CA, US (10G) IPv4 934 Mbps 1.04 Gbps
Leaseweb, NYC, NY, US (10G) IPv4 1.07 Gbps 1.07 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 5,389 0.186 ms
fdatasync 5,310 0.188 ms
fsync 2,886 0.347 ms
fsync_writethrough Not supported here n/a
open_sync 2,997 0.334 ms
Two 8 kB writes
open_datasync 2,802 0.357 ms
fdatasync 4,268 0.234 ms
fsync 2,361 0.424 ms
fsync_writethrough Not supported here n/a
open_sync 1,508 0.663 ms
open_sync at different write sizes
1 * 16kB open_sync write 1,645 0.608 ms
2 * 8kB open_sync writes 1,009 0.992 ms
4 * 4kB open_sync writes 521 1.92 ms
8 * 2kB open_sync writes 280 3.57 ms
16 * 1kB open_sync writes 139 7.2 ms
fsync on a non-write descriptor
write, fsync, close 1,695 0.59 ms
write, close, fsync 1,950 0.513 ms
Non-synced 8 kB writes
write 2,318,688 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 1,689 0.419 ms 1.28 ms 2.66 ms 14.6 ms 0
Read/write (TPC-B like) 8 5,741 1.03 ms 3.04 ms 6.8 ms 27.9 ms 0
Read/write (TPC-B like) 32 5,288 3.93 ms 17.5 ms 37.3 ms 84.8 ms 0
Read only 1 59,088 0.016 ms 0.024 ms 0.031 ms 0.099 ms 0
Read only 8 107,599 0.067 ms 0.116 ms 0.147 ms 0.25 ms 0
Read only 32 104,315 0.266 ms 0.448 ms 0.595 ms 1.16 ms 0

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 Failed
Read/write (TPC-B like) 1 1,578 0.507 ms 0.994 ms 2.31 ms 29.3 ms 0
Read/write (TPC-B like) 8 5,249 0.98 ms 2.99 ms 14.4 ms 62.9 ms 0
Read/write (TPC-B like) 32 6,633 3.41 ms 9.08 ms 44 ms 87.4 ms 0
Read only 1 13,599 0.033 ms 0.16 ms 0.391 ms 1.94 ms 0
Read only 8 67,616 0.084 ms 0.266 ms 0.444 ms 2.2 ms 0
Read only 32 65,322 0.393 ms 0.909 ms 1.8 ms 3.68 ms 0

Scale 100 (10,000,000 accounts), loaded in 14.2 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 1 389 7,783 2.03 ms 4.03 ms 14.2 ms 0
Read/write 8 1,098 21,955 5.67 ms 13 ms 56.8 ms 0
Read/write 32 1,389 27,777 20.7 ms 46.6 ms 74.5 ms 0
Read only 1 2,054 32,865 0.464 ms 0.702 ms 0.954 ms 0
Read only 8 4,151 66,411 1.82 ms 3.02 ms 3.55 ms 0
Read only 32 2,784 44,543 11 ms 17 ms 20 ms 0

10 tables of 100,000 rows, loaded in 7.4 s. 60 s per run, one connection per thread.

Redis

memtier

Mix Operations/s Hits/s Median p99 p99.9
1:10 set:get 167,590 152,354 0.215 ms 2.35 ms 4.1 ms
1:1 set:get 157,289 78,644 0.247 ms 2.35 ms 4.13 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 502 2,374 2,713 1.21 s
Express express-postgresjs 8,309 13,234 71,363 313 ms
Fastify fastify-postgres 8,411 14,795 83,944 55.3 ms
NestJS nestjs 8,776 10,811 67,551 82.5 ms
Nuxt nuxt 7,612 15,024 68,121 73.9 ms
FastAPI fastapi 6,344 9,359 32,608 151 ms
Django django-postgresql 2,500 3,434 9,706 202 ms
Rails rails 4,585 6,196 32,218 91 ms
Laravel laravel 1,033 1,206 2,010 317 ms
Symfony symfony 1,179 1,506 4,550 329 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 2.27 717 ms 865 ms 898 ms 0%
Shop page 2.27 964 ms 1.15 s 1.28 s 0%
Category page 2.27 582 ms 1.03 s 1.12 s 0%
Product page 2.27 642 ms 794 ms 857 ms 0%
Product search 2.27 541 ms 720 ms 771 ms 0%
32 virtual users
Home page 2.28 2.52 s 2.94 s 3.06 s 0%
Shop page 2.28 3.51 s 4.07 s 4.26 s 0%
Category page 2.28 2.97 s 3.74 s 3.95 s 0%
Product page 2.28 2.46 s 2.89 s 3 s 0%
Product search 2.28 2.32 s 2.69 s 2.82 s 0%

Cart and checkout

Step Requests/s Median p95 p99 Error rate
8 virtual users
Add to cart 4.75 700 ms 948 ms 1.04 s 0%
Cart 4.75 537 ms 706 ms 809 ms 0%
Checkout 4.75 419 ms 540 ms 605 ms 0%
32 virtual users
Add to cart 5.02 2.22 s 2.78 s 2.94 s 0%
Cart 5.02 2.29 s 2.68 s 2.84 s 0%
Checkout 5.02 1.78 s 2.23 s 2.29 s 0%

k6 with a fixed number of virtual users, 61 to 70 s per level. The error rate counts responses that weren't a 200.

Advertised vs. measured

Advertised Measured
vCPU 2 2
CPU Shared vCPU 2.7% 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
2.7% 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
New York #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 9, 2026, 23:21:27 UTC
Parser version
5

Run timeline

Queued
Oct 9, 2026, 23:20:00 UTC
Server created
Oct 9, 2026, 23:21:22 UTC
First update from the server
Oct 9, 2026, 23:21:22 UTC
Completed
Oct 10, 2026, 00:30:00 UTC
Server deleted
Oct 10, 2026, 00:30:00 UTC
Cost of this run
€0.031 EUR
Watch party 0 watching

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

Watch party

Loading the party