Skip to main content

RamNode 4GB Standard

Los Angeles, CA

Benchmark complete
Finished Oct 18, 2026
Geekbench 7 single-core
482
Faster than 0% of 4 vCPU plans
Geekbench 7 multi-core
Skipped
Disk 4K IOPS
30,587
fio random read and write, 50/50
Monthly price at test time
$20.00
Same price today

Server

  • Provider
    RamNode
  • Plan
    4GB Standard
  • Location
    Los Angeles, CA
  • vCPU
    4
  • RAM
    4 GB
  • Storage
    120 GB SSD
  • Network
    Unknown
  • OS image
    Ubuntu 24.04 Server Cloud
  • Trigger
    Scheduled
  • Queued at
    Oct 18, 2026, 02:55:00 UTC
Visit RamNode

CPU

Version Single-core Multi-core
Geekbench 7 482 Skipped Geekbench Browser

Multi-core was skipped because Geekbench 7 needs about 4.2 GB of free memory to run 4 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
375
239th of 241
Single-core only
Navigation
567
238th of 241
Single-core only
HTML5 Browser
487
235th of 241
Single-core only
PDF Viewer
533
237th of 241
Single-core only
Photo Library
413
233rd of 241
Single-core only
Clang
407
237th of 241
Single-core only
Text Processing
390
232nd of 241
Single-core only
Asset Compression
523
234th of 241
Single-core only
HDR
488
237th of 241
Single-core only
Photo Editor
521
229th of 241
Single-core only
Ray Tracer
570
225th of 241
Single-core only
Structure from Motion
578
236th of 241
Single-core only
Game Physics
432
237th of 241
Single-core only
Video Encoder
466
234th of 241
Single-core only
Audio Encoder
572
234th of 241
Single-core only
Video Player
461
234th 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 17.3 GB/s 3 11.8%
Scale higher is better 9.59 GB/s 3 12.4%
Add higher is better 10.2 GB/s 3 6.06%
Triad higher is better 10 GB/s 3 12.3%

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 225/s 3 6.75%
RSA 4096 verifies higher is better 15,695/s 3 2.96%
SHA-256 higher is better 432 MB/s 3 7.55%
AES-256-GCM higher is better 3.37 GB/s 3 23%
ChaCha20-Poly1305 higher is better 3.47 GB/s 3 4.91%

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 9.45 ms 3 1.74%
Requests per second, 90th percentile higher is better 425/s 3 2.82%
Request latency, 99.9th percentile lower is better 87.6 ms 3 27%
Requests per second, average higher is better 326/s 3 3.83%

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

Disk

Block size Read Write Total
4K 59.7 MB/s (15,279 IOPS) 59.8 MB/s (15,308 IOPS) 119.5 MB/s (30,587 IOPS)
64K 257.8 MB/s (4,124 IOPS) 259.2 MB/s (4,146 IOPS) 517 MB/s (8,270 IOPS)
512K 224.8 MB/s (449 IOPS) 236.8 MB/s (473 IOPS) 461.6 MB/s (922 IOPS)
1M 245.2 MB/s (245 IOPS) 261.5 MB/s (261 IOPS) 506.7 MB/s (506 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 1.22 Gbps 1.03 Gbps
Eranium, Amsterdam, NL (100G) IPv4 1.27 Gbps 1.38 Gbps
Leaseweb, Singapore, SG (10G) IPv4 399 Mbps 1.19 Gbps
Clouvider, Los Angeles, CA, US (10G) IPv4 5.54 Gbps 3.17 Gbps
Leaseweb, NYC, NY, US (10G) IPv4 2.48 Gbps 614 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 929 1.08 ms
fdatasync 1,090 0.918 ms
fsync 388 2.58 ms
fsync_writethrough Not supported here n/a
open_sync 400 2.5 ms
Two 8 kB writes
open_datasync 589 1.7 ms
fdatasync 1,112 0.9 ms
fsync 428 2.34 ms
fsync_writethrough Not supported here n/a
open_sync 188 5.31 ms
open_sync at different write sizes
1 * 16kB open_sync write 382 2.62 ms
2 * 8kB open_sync writes 199 5.04 ms
4 * 4kB open_sync writes 99.8 10 ms
8 * 2kB open_sync writes 52.1 19.2 ms
16 * 1kB open_sync writes 26.3 38.1 ms
fsync on a non-write descriptor
write, fsync, close 402 2.49 ms
write, close, fsync 412 2.43 ms
Non-synced 8 kB writes
write 403,537 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) 1 246 2.59 ms 12.1 ms 22.5 ms 55.9 ms 0
Read/write (TPC-B like) 8 638 8.06 ms 35 ms 57.5 ms 95.9 ms 0
Read/write (TPC-B like) 32 772 27.2 ms 126 ms 238 ms 502 ms 0
Read only 1 4,639 0.126 ms 0.452 ms 1.56 ms 8.46 ms 0
Read only 8 22,108 0.189 ms 0.61 ms 3.88 ms 18.8 ms 0
Read only 32 22,004 0.647 ms 4.12 ms 18 ms 51.1 ms 0

Scale 10 (1,000,000 accounts), loaded in 4.76 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) 1 212 3.02 ms 13.7 ms 24.9 ms 50.4 ms 0
Read/write (TPC-B like) 8 625 8.16 ms 36.8 ms 70.8 ms 153 ms 0
Read/write (TPC-B like) 32 953 23.3 ms 89.3 ms 144 ms 257 ms 0
Read only 1 1,521 0.289 ms 2.22 ms 7.28 ms 15.1 ms 0
Read only 8 14,466 0.247 ms 1.3 ms 7.76 ms 25.4 ms 0
Read only 32 16,866 0.872 ms 5.37 ms 23.8 ms 70.1 ms 0

Scale 100 (10,000,000 accounts), loaded in 44.9 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.

MySQL

sysbench OLTP

Workload Threads Transactions/s Queries/s Median p95 p99 Errors
Read/write 1 38.2 763 21.1 ms 60 ms 101 ms 0
Read/write 8 184 3,673 33.7 ms 111 ms 177 ms 0
Read/write 32 322 6,434 78.6 ms 227 ms 397 ms 0
Read only 1 130 2,077 5.57 ms 19 ms 33.7 ms 0
Read only 8 706 11,293 6.67 ms 35 ms 86 ms 0
Read only 32 768 12,287 31.4 ms 107 ms 204 ms 0

10 tables of 100,000 rows, loaded in 28.4 s. 60 s per run, one connection per thread.

Redis

memtier

Mix Operations/s Hits/s Median p99 p99.9
1:10 set:get 15,694 14,267 1.81 ms 29.8 ms 60.2 ms
1:1 set:get 17,778 8,889 1.77 ms 25 ms 49.2 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 242 819 1,148 3.47 s
Express express-postgresjs 4,179 4,636 10,091 340 ms
Fastify fastify-postgres 4,917 5,354 10,910 192 ms
NestJS nestjs 1,180 1,894 6,752 1.59 s
Nuxt nuxt 2,975 3,835 6,752 470 ms
FastAPI fastapi 1,644 2,119 5,626 786 ms
Django django-postgresql 446 535 1,008 1.5 s
Rails rails 672 1,109 3,116 1.46 s
Laravel laravel 239 286 536 1.83 s
Symfony symfony 230 318 758 2.02 s

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 0.589 2.2 s 3.59 s 3.87 s 0%
Shop page 0.589 3.11 s 4.28 s 4.46 s 0%
Category page 0.589 2.19 s 3.46 s 5.54 s 0%
Product page 0.589 2.24 s 3.52 s 4.34 s 0%
Product search 0.589 1.59 s 2.71 s 3.05 s 0%
32 virtual users
Home page 0.859 6.04 s 8.28 s 8.61 s 0%
Shop page 0.859 8.42 s 10.2 s 10.7 s 0%
Category page 0.859 7.57 s 9.49 s 9.96 s 0%
Product page 0.832 7.26 s 9.1 s 10.3 s 0%
Product search 0.55 5.84 s 7.79 s 8.91 s 0%

Cart and checkout

Step Requests/s Median p95 p99 Error rate
8 virtual users
Add to cart 1.49 2 s 2.88 s 3.36 s 0%
Cart 1.49 1.76 s 2.37 s 2.71 s 0%
Checkout 1.49 1.27 s 2.09 s 2.56 s 0%
32 virtual users
Add to cart 1.97 5.34 s 6.88 s 7.12 s 0%
Cart 1.97 5.43 s 6.57 s 6.77 s 0%
Checkout 1.97 4.26 s 5.78 s 6.06 s 0%

k6 with a fixed number of virtual users, 65 to 74 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 0.216 1.99 s 2.96 s 3.45 s 0%
Topic JSON 0.108 962 ms 1.67 s 1.7 s 0%
Home page 0.108 1.95 s 2.38 s 2.4 s 0%
Categories page 0.0617 1.86 s 2.67 s 2.77 s 0%
User activity page 0.0463 1.01 s 1.31 s 1.33 s 0%
4 virtual users
Topic page 0.53 2.89 s 4.14 s 6.05 s 0%
Topic JSON 0.28 1.61 s 2.49 s 2.97 s 0%
Home page 0.28 2.26 s 5.93 s 7.3 s 0%
Categories page 0.14 2.76 s 3.55 s 3.6 s 0%
User activity page 0.125 2.6 s 4.3 s 4.83 s 0%
16 virtual users
Topic page 0.512 7.48 s 10.8 s 12.5 s 0%
Topic JSON 0.263 4.06 s 6.32 s 6.35 s 0%
Home page 0.263 6.93 s 9.02 s 10.6 s 0%
Categories page 0.132 7.2 s 12.5 s 13.2 s 0%
User activity page 0.161 4.47 s 10.3 s 11.8 s 0%

Anonymous

Step Requests/s Median p95 p99 Error rate
16 virtual users
Topic page 11.5 282 ms 812 ms 1.31 s 2.48%
Topic JSON 5.84 268 ms 788 ms 1.05 s 2.72%
Home page 5.79 282 ms 789 ms 2.1 s 4.11%
Categories page 2.92 276 ms 914 ms 3.98 s 1.63%
User activity page 2.93 1.49 s 3.37 s 4.31 s 1.62%

k6 with a fixed number of virtual users, 63 to 68 s per level, after a 32 s warmup. The error rate counts responses that weren't a 200. Discourse ran with 8 app workers and 1024 MB of Postgres shared buffers, what its installer picks for a 4 GB box. It added a 2 GB swapfile first, the way its installer does, and didn't need it. The container peaked at 2,790 MB.

Ghost

Readers

Step Requests/s Median p95 p99 Error rate
1 virtual user
Home page 0.924 284 ms 473 ms 582 ms 0%
Post page 2.26 159 ms 353 ms 576 ms 0%
Tag page 0.462 185 ms 412 ms 566 ms 0%
Author page 0.446 191 ms 382 ms 425 ms 0%
RSS feed 0.446 210 ms 322 ms 470 ms 0%
4 virtual users
Home page 1.14 886 ms 1.42 s 1.66 s 0%
Post page 2.85 682 ms 1.08 s 1.39 s 0%
Tag page 0.577 506 ms 866 ms 1.02 s 0%
Author page 0.561 474 ms 748 ms 957 ms 0%
RSS feed 0.544 644 ms 854 ms 928 ms 0%
16 virtual users
Home page 1.28 3.03 s 4.59 s 4.87 s 0%
Post page 3.18 2.59 s 4 s 4.46 s 0%
Tag page 0.632 1.58 s 2.29 s 2.63 s 0%
Author page 0.632 1.64 s 2.62 s 2.9 s 0%
RSS feed 0.6 1.76 s 3.04 s 3.39 s 0%

Staff in Ghost Admin

Step Requests/s Median p95 p99 Error rate
1 virtual user
Post list 0.39 266 ms 617 ms 674 ms 0%
Post editor 1.2 151 ms 338 ms 407 ms 0%
Saving a draft 1.2 373 ms 592 ms 917 ms 0%
Member list 0.39 387 ms 955 ms 1.06 s 0%
4 virtual users
Post list 0.536 367 ms 664 ms 769 ms 0%
Post editor 1.71 211 ms 463 ms 569 ms 0%
Saving a draft 1.72 1.59 s 2.58 s 3.08 s 0%
Member list 0.585 454 ms 837 ms 1.17 s 0%

k6 with a fixed number of virtual users, 61 to 62 s per level, after a 21 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 271 MB and MySQL at 626 MB.

Wagtail

Visitors

Step Requests/s Median p95 p99 Error rate
1 virtual user
Home page 0.207 585 ms 774 ms 777 ms 0%
Blog index 0.207 2.44 s 3.64 s 4.22 s 0%
Blog post 0.589 163 ms 280 ms 313 ms 0%
Bread page 0.605 191 ms 316 ms 408 ms 0%
Location page 0.207 172 ms 337 ms 365 ms 0%
Site search 0.191 343 ms 639 ms 674 ms 0%
4 virtual users
Home page 0.458 1.13 s 1.62 s 1.7 s 0%
Blog index 0.458 4.42 s 7.28 s 7.6 s 0%
Blog post 1.33 313 ms 585 ms 635 ms 0%
Bread page 1.36 361 ms 511 ms 585 ms 0%
Location page 0.458 327 ms 507 ms 520 ms 0%
Site search 0.442 451 ms 1.03 s 1.15 s 0%
16 virtual users
Home page 0.522 3.55 s 4.75 s 4.86 s 0%
Blog index 0.552 8.66 s 11.1 s 11.5 s 0%
Blog post 1.5 1.89 s 3.43 s 3.89 s 0%
Bread page 1.53 2 s 3.41 s 3.64 s 0%
Location page 0.522 1.88 s 3.24 s 3.53 s 0%
Site search 0.491 2.3 s 3.83 s 4.71 s 0%

Editors in the Wagtail admin

Step Requests/s Median p95 p99 Error rate
1 virtual user
Admin dashboard 0.0898 620 ms 743 ms 744 ms 0%
Page explorer 0.195 2.08 s 2.8 s 3.23 s 0%
Page editor 0.269 1.06 s 1.59 s 1.6 s 0%
Image library 0.0898 546 ms 3.3 s 3.92 s 0%
Admin page search 0.0898 362 ms 670 ms 723 ms 0%
4 virtual users
Admin dashboard 0.2 1.12 s 2.44 s 2.73 s 0%
Page explorer 0.4 4.03 s 7.01 s 7.55 s 0%
Page editor 0.6 2.29 s 3.48 s 4.34 s 0%
Image library 0.2 892 ms 2.01 s 2.5 s 0%
Admin page search 0.2 570 ms 1.26 s 1.35 s 0%

k6 with a fixed number of virtual users, 61 to 67 s per level, after a 22 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 285 MB, Postgres at 211 MB, and Redis at 51 MB.

Advertised vs. measured

Advertised Measured
vCPU 4 4
CPU Shared vCPU 36.2% of busy CPU time stolen
RAM 4 GB 3.8 GiB
Storage 120 GB SSD 116.2 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
Intel Core Processor (Haswell, no TSX, IBRS)
CPU cores
4
CPU frequency
2300 MHz
AES-NI
Yes
VM-x/AMD-V
Yes
RAM reported
3.8 GiB
Swap reported
0 MiB
Disk reported
116.2 GiB
Virtualization reported
KVM
CPU steal
36.2% of busy CPU time stolen
Distro
Ubuntu 24.04.1 LTS
Kernel
6.8.0-51-generic
Architecture
x86_64
IPv4 reachable
Yes
IPv6 reachable
No

Advertised at test time

Plan
4GB Standard
Product line
Standard VPS
Location
Los Angeles, CA
vCPU
4 shared vCPU
Architecture
x86_64
RAM
4 GB
Storage
120 GB SSD
Virtualization
KVM
Network
Unknown
Transfer
8 TB
Port 25
Unknown
Monthly price at test time
$20.00
Hourly price at test time
$0.030
YABS version
v2026-09-20
Ran at
Oct 18, 2026, 02:56:58 UTC
Parser version
5

Run timeline

Queued
Oct 18, 2026, 02:55:00 UTC
Server created
Oct 18, 2026, 02:56:14 UTC
First update from the server
Oct 18, 2026, 02:56:14 UTC
Completed
Oct 18, 2026, 05:17:00 UTC
Server deleted
Oct 18, 2026, 05:17:00 UTC
Cost of this run
$0.071
Watch party 0 watching

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

Watch party

Loading the party