Skip to main content

Vultr High Performance AMD 2 vCPU / 2 GB

Chicago, US

Benchmark complete
Finished Oct 17, 2026
Geekbench 7 single-core
1,873
Faster than 89% of 2 vCPU plans
Geekbench 7 multi-core
Skipped
Disk 4K IOPS
179,060
fio random read and write, 50/50
Monthly price at test time
$18.00
Same price today

Server

  • Provider
    Vultr
  • Plan
    High Performance AMD 2 vCPU / 2 GB
  • Location
    Chicago, US
  • vCPU
    2
  • RAM
    2 GB
  • Storage
    60 GB NVMe
  • Network
    Unknown
  • OS image
    2284
  • Trigger
    Scheduled
  • Queued at
    Oct 17, 2026, 01:24:00 UTC
Visit Vultr

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 1,873 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
1,936
11th of 216
Single-core only
Navigation
1,845
15th of 216
Single-core only
HTML5 Browser
1,908
19th of 216
Single-core only
PDF Viewer
1,929
24th of 216
Single-core only
Photo Library
1,955
13th of 216
Single-core only
Clang
1,753
10th of 216
Single-core only
Text Processing
1,793
20th of 216
Single-core only
Asset Compression
1,838
25th of 216
Single-core only
HDR
1,980
15th of 216
Single-core only
Photo Editor
2,218
18th of 216
Single-core only
Ray Tracer
1,710
19th of 216
Single-core only
Structure from Motion
1,810
30th of 216
Single-core only
Game Physics
1,804
16th of 216
Single-core only
Video Encoder
1,753
27th of 216
Single-core only
Audio Encoder
1,898
13th of 216
Single-core only
Video Player
1,889
26th of 216
Single-core only

Rank among the 216 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 56.2 GB/s 3 3.52%
Scale higher is better 70.9 GB/s 3 4.26%
Add higher is better 52.2 GB/s 3 0.887%
Triad higher is better 52.2 GB/s 3 0.886%

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 1,449/s 3 0.348%
RSA 4096 verifies higher is better 25,128/s 3 0.404%
SHA-256 higher is better 2.19 GB/s 3 0.844%
AES-256-GCM higher is better 22.1 GB/s 3 0.309%
ChaCha20-Poly1305 higher is better 8.29 GB/s 3 0.197%

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 3.73 ms 3 0.248%
Requests per second, 90th percentile higher is better 504/s 3 0.303%
Request latency, 99.9th percentile lower is better 8.17 ms 3 1.09%
Requests per second, average higher is better 496/s 3 0.551%

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

Disk

Block size Read Write Total
4K 349.3 MB/s (89,412 IOPS) 350.2 MB/s (89,648 IOPS) 699.5 MB/s (179,060 IOPS)
64K 1.21 GB/s (19,903 IOPS) 1.22 GB/s (20,008 IOPS) 2.44 GB/s (39,911 IOPS)
512K 1.41 GB/s (2,897 IOPS) 1.49 GB/s (3,051 IOPS) 2.9 GB/s (5,948 IOPS)
1M 1.57 GB/s (1,606 IOPS) 1.67 GB/s (1,713 IOPS) 3.24 GB/s (3,319 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 1.77 Gbps 2.04 Gbps
Eranium, Amsterdam, NL (100G) IPv4 8.04 Gbps 6.35 Gbps
Leaseweb, Singapore, SG (10G) IPv4 2.91 Gbps 5.15 Gbps
Clouvider, Los Angeles, CA, US (10G) IPv4 3.81 Gbps 4.65 Gbps
Leaseweb, NYC, NY, US (10G) IPv4 6.4 Gbps 9.3 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 12,744 0.078 ms
fdatasync 12,407 0.081 ms
fsync 9,730 0.103 ms
fsync_writethrough Not supported here n/a
open_sync 10,529 0.095 ms
Two 8 kB writes
open_datasync 5,953 0.168 ms
fdatasync 11,111 0.09 ms
fsync 9,467 0.106 ms
fsync_writethrough Not supported here n/a
open_sync 5,507 0.182 ms
open_sync at different write sizes
1 * 16kB open_sync write 10,680 0.094 ms
2 * 8kB open_sync writes 5,706 0.175 ms
4 * 4kB open_sync writes 2,944 0.34 ms
8 * 2kB open_sync writes 1,398 0.715 ms
16 * 1kB open_sync writes 708 1.41 ms
fsync on a non-write descriptor
write, fsync, close 8,868 0.113 ms
write, close, fsync 9,002 0.111 ms
Non-synced 8 kB writes
write 591,614 0.002 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 2,741 0.356 ms 0.435 ms 0.52 ms 0.959 ms 0
Read/write (TPC-B like) 8 6,254 1.2 ms 2.04 ms 2.75 ms 4.42 ms 0
Read/write (TPC-B like) 32 5,324 4.8 ms 13.7 ms 22.7 ms 37.3 ms 0
Read only 1 29,030 0.032 ms 0.045 ms 0.056 ms 0.094 ms 0
Read only 8 65,208 0.105 ms 0.181 ms 0.23 ms 0.412 ms 0
Read only 32 63,676 0.378 ms 0.759 ms 1.09 ms 2.16 ms 0

Scale 10 (1,000,000 accounts), loaded in 0.93 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 1,826 0.527 ms 0.721 ms 0.996 ms 2.64 ms 0
Read/write (TPC-B like) 8 4,857 1.56 ms 2.49 ms 3.7 ms 6.36 ms 0
Read/write (TPC-B like) 32 5,530 5.48 ms 8.94 ms 11.4 ms 14.8 ms 0
Read only 1 8,946 0.095 ms 0.205 ms 0.348 ms 1.18 ms 0
Read only 8 41,944 0.147 ms 0.395 ms 0.556 ms 2.1 ms 0
Read only 32 42,135 0.617 ms 1.2 ms 1.92 ms 3.59 ms 0

Scale 100 (10,000,000 accounts), loaded in 9.86 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 1 552 11,043 1.73 ms 2.35 ms 3.82 ms 0
Read/write 8 1,010 20,206 7.84 ms 12.1 ms 15.3 ms 0
Read/write 32 1,140 22,795 27.7 ms 45 ms 56.8 ms 0
Read only 1 1,184 18,938 0.826 ms 1.03 ms 1.18 ms 0
Read only 8 2,353 37,654 3.19 ms 5.09 ms 5.88 ms 0
Read only 32 2,131 34,093 14.5 ms 20 ms 22.3 ms 0

10 tables of 100,000 rows, loaded in 5 s. 60 s per run, one connection per thread.

Redis

memtier

Mix Operations/s Hits/s Median p99 p99.9
1:10 set:get 67,677 61,524 0.727 ms 2.67 ms 3.34 ms
1:1 set:get 67,049 33,524 0.735 ms 2.72 ms 3.49 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 796 2,832 3,992 458 ms
Express express-postgresjs 8,846 12,706 49,137 51.4 ms
Fastify fastify-postgres 8,643 13,060 51,221 52.7 ms
NestJS nestjs 6,121 7,140 27,126 79 ms
Nuxt nuxt 7,911 12,765 43,874 64.5 ms
FastAPI fastapi 4,794 5,823 28,987 294 ms
Django django-postgresql 2,362 2,656 6,584 168 ms
Rails rails 3,701 5,146 19,000 107 ms
Laravel laravel 1,012 1,044 2,001 270 ms
Symfony symfony 1,289 1,382 3,832 210 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 3.08 522 ms 600 ms 626 ms 0%
Shop page 3.08 683 ms 782 ms 800 ms 0%
Category page 3.08 437 ms 703 ms 774 ms 0%
Product page 3.08 482 ms 562 ms 571 ms 0%
Product search 3.08 416 ms 497 ms 512 ms 0%
32 virtual users
Home page 3.05 1.92 s 2.17 s 2.24 s 0%
Shop page 3.05 2.55 s 2.89 s 2.97 s 0%
Category page 3.05 2.22 s 2.67 s 2.93 s 0%
Product page 3.05 1.93 s 2.09 s 2.14 s 0%
Product search 3.05 1.79 s 1.96 s 2.06 s 0%

Cart and checkout

Step Requests/s Median p95 p99 Error rate
8 virtual users
Add to cart 5.96 567 ms 672 ms 698 ms 0%
Cart 5.96 458 ms 551 ms 577 ms 0%
Checkout 5.96 356 ms 431 ms 459 ms 0%
32 virtual users
Add to cart 6 1.82 s 2.21 s 2.28 s 0%
Cart 6 1.92 s 2.13 s 2.18 s 0%
Checkout 6 1.53 s 1.76 s 1.82 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 2.94 150 ms 227 ms 395 ms 0%
Topic JSON 1.48 76.1 ms 93.8 ms 247 ms 0%
Home page 1.48 134 ms 190 ms 291 ms 0%
Categories page 0.73 153 ms 167 ms 177 ms 0%
User activity page 0.73 92.7 ms 177 ms 255 ms 0%
4 virtual users
Topic page 5.21 341 ms 464 ms 820 ms 0%
Topic JSON 2.63 177 ms 225 ms 243 ms 0%
Home page 2.62 301 ms 451 ms 841 ms 0%
Categories page 1.32 347 ms 473 ms 552 ms 0%
User activity page 1.32 215 ms 483 ms 627 ms 0%
16 virtual users
Topic page 5.22 1.22 s 1.57 s 1.75 s 0%
Topic JSON 2.63 1.07 s 1.29 s 1.35 s 0%
Home page 2.61 1.19 s 1.48 s 1.54 s 0%
Categories page 1.29 1.21 s 1.55 s 1.88 s 0%
User activity page 1.3 1.13 s 1.33 s 1.37 s 0%

Anonymous

Step Requests/s Median p95 p99 Error rate
16 virtual users
Topic page 103 43.6 ms 108 ms 143 ms 0%
Topic JSON 51.7 41.4 ms 107 ms 143 ms 0%
Home page 51.7 42.5 ms 103 ms 143 ms 0%
Categories page 25.8 42.9 ms 107 ms 140 ms 0%
User activity page 25.8 139 ms 225 ms 330 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 4 app workers and 256 MB of Postgres shared buffers, what its installer picks for a 2 GB box. It had swap the way its installer sets a box up and used up to 409 MB of it. The container peaked at 1,450 MB.

Ghost

Readers

Step Requests/s Median p95 p99 Error rate
1 virtual user
Home page 11.7 23.3 ms 29.6 ms 34.8 ms 0%
Post page 29.3 13 ms 16.7 ms 24.3 ms 0%
Tag page 5.86 16.2 ms 20.1 ms 29.8 ms 0%
Author page 5.86 16 ms 20.1 ms 27 ms 0%
RSS feed 5.85 19.9 ms 25 ms 36.1 ms 0%
4 virtual users
Home page 14 72.4 ms 97.4 ms 129 ms 0%
Post page 35 53.1 ms 78.3 ms 103 ms 0%
Tag page 7.01 47.7 ms 72.2 ms 96.4 ms 0%
Author page 6.99 45.3 ms 65.8 ms 94.8 ms 0%
RSS feed 6.99 51.4 ms 70.5 ms 96.1 ms 0%
16 virtual users
Home page 13.7 300 ms 379 ms 403 ms 0%
Post page 34.2 238 ms 319 ms 349 ms 0%
Tag page 6.83 170 ms 233 ms 261 ms 0%
Author page 6.86 166 ms 238 ms 258 ms 0%
RSS feed 6.85 167 ms 243 ms 267 ms 0%

Staff in Ghost Admin

Step Requests/s Median p95 p99 Error rate
1 virtual user
Post list 6.7 22.2 ms 30.6 ms 45 ms 0%
Post editor 20.1 8.87 ms 12.3 ms 17.5 ms 0%
Saving a draft 20.1 21.5 ms 27.9 ms 40 ms 0%
Member list 6.7 25.2 ms 32.2 ms 36.6 ms 0%
4 virtual users
Post list 8.01 31.4 ms 41 ms 56.5 ms 0%
Post editor 24.1 16.1 ms 21.7 ms 34.6 ms 0%
Saving a draft 24.1 121 ms 162 ms 198 ms 0%
Member list 8.04 33.1 ms 42.9 ms 63.4 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 369 MB and MySQL at 583 MB.

Wagtail

Visitors

Step Requests/s Median p95 p99 Error rate
1 virtual user
Home page 2.44 54.3 ms 61.5 ms 65.2 ms 0%
Blog index 2.44 183 ms 196 ms 209 ms 0%
Blog post 7.3 17.5 ms 26.9 ms 28.9 ms 0%
Bread page 7.31 19.2 ms 22.3 ms 24.8 ms 0%
Location page 2.44 18.5 ms 22.9 ms 25.3 ms 0%
Site search 2.43 29.7 ms 50.6 ms 53 ms 0%
4 virtual users
Home page 3.85 146 ms 215 ms 255 ms 0%
Blog index 3.88 509 ms 661 ms 727 ms 0%
Blog post 11.6 42.1 ms 71.6 ms 97.4 ms 0%
Bread page 11.6 45.9 ms 68.5 ms 76.7 ms 0%
Location page 3.88 44.4 ms 63.7 ms 74.7 ms 0%
Site search 3.85 67 ms 144 ms 175 ms 0%
16 virtual users
Home page 4.38 419 ms 570 ms 672 ms 0%
Blog index 4.36 1.05 s 1.23 s 1.33 s 0%
Blog post 13 240 ms 408 ms 558 ms 0%
Bread page 13.1 249 ms 405 ms 479 ms 0%
Location page 4.36 244 ms 424 ms 554 ms 0%
Site search 4.36 305 ms 467 ms 531 ms 0%

Editors in the Wagtail admin

Step Requests/s Median p95 p99 Error rate
1 virtual user
Admin dashboard 0.975 56.9 ms 66.1 ms 91.2 ms 0%
Page explorer 1.97 243 ms 276 ms 385 ms 0%
Page editor 2.93 117 ms 139 ms 218 ms 0%
Image library 0.975 44.7 ms 49.4 ms 301 ms 0%
Admin page search 0.975 25.7 ms 51.5 ms 53.1 ms 0%
4 virtual users
Admin dashboard 1.56 153 ms 388 ms 497 ms 0%
Page explorer 3.15 616 ms 936 ms 1.08 s 0%
Page editor 4.68 277 ms 539 ms 683 ms 0%
Image library 1.56 137 ms 417 ms 433 ms 0%
Admin page search 1.56 76.5 ms 218 ms 277 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 302 MB, Postgres at 175 MB, and Redis at 26 MB.

Advertised vs. measured

Advertised Measured
vCPU 2 2
CPU Shared vCPU 0% of busy CPU time stolen
RAM 2 GB 1.9 GiB
Storage 60 GB NVMe 56.2 GiB
Architecture x86_64 x86_64
Virtualization KVM MICROSOFT

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
2900 MHz
AES-NI
Yes
VM-x/AMD-V
No
RAM reported
1.9 GiB
Swap reported
5.8 GiB
Disk reported
56.2 GiB
Virtualization reported
MICROSOFT
CPU steal
0% of busy CPU time stolen
Distro
Ubuntu 24.04.5 LTS
Kernel
6.8.0-146-generic
Architecture
x86_64
IPv4 reachable
Yes
IPv6 reachable
Yes

Advertised at test time

Plan
High Performance AMD 2 vCPU / 2 GB
Product line
Cloud Compute High Performance
Location
Chicago, US
vCPU
2 shared vCPU
Architecture
x86_64
RAM
2 GB
Storage
60 GB NVMe
Virtualization
KVM
Network
Unknown
Transfer
4.1 TB
Port 25
On request
Monthly price at test time
$18.00
Hourly price at test time
$0.025
YABS version
v2026-09-20
Ran at
Oct 17, 2026, 01:25:46 UTC
Parser version
5

Run timeline

Queued
Oct 17, 2026, 01:24:00 UTC
Server created
Oct 17, 2026, 01:24:50 UTC
First update from the server
Oct 17, 2026, 01:24:50 UTC
Completed
Oct 17, 2026, 02:52:04 UTC
Server deleted
Oct 17, 2026, 02:52:04 UTC
Cost of this run
$0.037
Watch party 0 watching

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

Watch party

Loading the party