Skip to main content

RamNode 3GB Standard

Los Angeles, CA

Benchmark complete
Finished Oct 17, 2026
Geekbench 7 single-core
254
Faster than 0% of 2 vCPU plans
Geekbench 7 multi-core
Skipped
Disk 4K IOPS
13,792
fio random read and write, 50/50
Monthly price at test time
$15.00
Same price today

Server

  • Provider
    RamNode
  • Plan
    3GB Standard
  • Location
    Los Angeles, CA
  • vCPU
    2
  • RAM
    3 GB
  • Storage
    100 GB SSD
  • Network
    Unknown
  • OS image
    Ubuntu 24.04 Server Cloud
  • Trigger
    Scheduled
  • Queued at
    Oct 17, 2026, 10:52:00 UTC
Visit RamNode

CPU

Version Single-core Multi-core
Geekbench 7 254 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
315
241st of 241
Single-core only
Navigation
380
241st of 241
Single-core only
HTML5 Browser
229
241st of 241
Single-core only
PDF Viewer
271
241st of 241
Single-core only
Photo Library
238
239th of 241
Single-core only
Clang
216
241st of 241
Single-core only
Text Processing
195
241st of 241
Single-core only
Asset Compression
273
241st of 241
Single-core only
HDR
270
241st of 241
Single-core only
Photo Editor
296
241st of 241
Single-core only
Ray Tracer
246
241st of 241
Single-core only
Structure from Motion
280
241st of 241
Single-core only
Game Physics
204
241st of 241
Single-core only
Video Encoder
209
241st of 241
Single-core only
Audio Encoder
309
241st of 241
Single-core only
Video Player
204
241st 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 20.3 GB/s 3 23.3%
Scale higher is better 10.8 GB/s 3 27.1%
Add higher is better 11.1 GB/s 3 25.6%
Triad higher is better 11.6 GB/s 3 34.9%

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 74.4/s 3 16%
RSA 4096 verifies higher is better 5,277/s 3 8.47%
SHA-256 higher is better 123 MB/s 3 9.71%
AES-256-GCM higher is better 1.57 GB/s 3 7.58%
ChaCha20-Poly1305 higher is better 1.13 GB/s 3 14.6%

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 13.4 ms 3 9.99%
Requests per second, 90th percentile higher is better 137/s 3 5.37%
Request latency, 99.9th percentile lower is better 105 ms 3 6.56%
Requests per second, average higher is better 111/s 3 10.9%

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

Disk

Block size Read Write Total
4K 26.9 MB/s (6,893 IOPS) 27 MB/s (6,899 IOPS) 53.9 MB/s (13,792 IOPS)
64K 259.8 MB/s (4,157 IOPS) 261.2 MB/s (4,179 IOPS) 521.1 MB/s (8,336 IOPS)
512K 203.8 MB/s (407 IOPS) 214.6 MB/s (429 IOPS) 418.4 MB/s (836 IOPS)
1M 235.1 MB/s (235 IOPS) 250.8 MB/s (250 IOPS) 485.9 MB/s (485 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 409 Mbps 809 Mbps
Eranium, Amsterdam, NL (100G) IPv4 1.14 Gbps 1.47 Gbps
Leaseweb, Singapore, SG (10G) IPv4 890 Mbps 1.36 Gbps
Clouvider, Los Angeles, CA, US (10G) IPv4 5.38 Gbps 3.11 Gbps
Leaseweb, NYC, NY, US (10G) IPv4 1.19 Gbps 2.22 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 470 2.13 ms
fdatasync 419 2.39 ms
fsync 170 5.9 ms
fsync_writethrough Not supported here n/a
open_sync 171 5.85 ms
Two 8 kB writes
open_datasync 254 3.93 ms
fdatasync 473 2.12 ms
fsync 180 5.57 ms
fsync_writethrough Not supported here n/a
open_sync 89.7 11.1 ms
open_sync at different write sizes
1 * 16kB open_sync write 187 5.34 ms
2 * 8kB open_sync writes 114 8.74 ms
4 * 4kB open_sync writes 52.8 18.9 ms
8 * 2kB open_sync writes 22.8 43.9 ms
16 * 1kB open_sync writes 11.8 85.1 ms
fsync on a non-write descriptor
write, fsync, close 187 5.36 ms
write, close, fsync 181 5.53 ms
Non-synced 8 kB writes
write 303,427 0.003 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) 1 145 4.58 ms 19.6 ms 32.1 ms n/a 0
Read/write (TPC-B like) 8 347 16.3 ms 64.7 ms 113 ms 247 ms 0
Read/write (TPC-B like) 32 333 69.2 ms 269 ms 457 ms 828 ms 0
Read only 1 1,803 0.235 ms 1.71 ms 6.33 ms 18.3 ms 0
Read only 8 7,348 0.413 ms 3.19 ms 14.7 ms 43.1 ms 0
Read only 32 6,600 1.7 ms 17.2 ms 43.5 ms 92.5 ms 0

Scale 10 (1,000,000 accounts), loaded in 7.09 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) 1 120 6.1 ms 24.6 ms 39 ms n/a 0
Read/write (TPC-B like) 8 356 15.8 ms 58.8 ms 98.2 ms 146 ms 0
Read/write (TPC-B like) 32 411 60.1 ms 177 ms 269 ms 462 ms 0
Read only 1 496 0.978 ms 7.01 ms 15.2 ms 35.7 ms 0
Read only 8 1,803 1.66 ms 17.8 ms 35.2 ms 67.9 ms 0
Read only 32 2,908 4.36 ms 39 ms 75.8 ms 138 ms 0

Scale 100 (10,000,000 accounts), loaded in 83.6 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 25.2 504 33.1 ms 89.2 ms 145 ms 0
Read/write 8 84.6 1,693 78.6 ms 204 ms 344 ms 0
Read/write 32 115 2,304 249 ms 580 ms 802 ms 0
Read only 1 73 1,168 11.2 ms 32.5 ms 51 ms 0
Read only 8 254 4,061 21.9 ms 89.2 ms 156 ms 0
Read only 32 298 4,768 90.8 ms 236 ms 350 ms 0

10 tables of 100,000 rows, loaded in 57.5 s. 60 to 61 s per run, one connection per thread.

Redis

memtier

Mix Operations/s Hits/s Median p99 p99.9
1:10 set:get 9,285 8,440 4.26 ms 39.4 ms 80.4 ms
1:1 set:get 8,089 4,044 4.7 ms 48.1 ms 105 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 65.3 248 376 7.35 s
Express express-postgresjs 2,040 1,719 3,565 2.07 s
Fastify fastify-postgres 1,468 2,644 3,983 3.17 s
NestJS nestjs 684 713 2,634 1.39 s
Nuxt nuxt 1,394 1,501 3,182 2.43 s
FastAPI fastapi 582 865 2,535 1.68 s
Django django-postgresql 182 202 459 2.59 s
Rails rails 320 411 1,485 1.51 s
Laravel laravel 128 143 266 2.51 s
Symfony symfony 166 190 440 2.18 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 0.369 4.02 s 5.1 s 5.19 s 0%
Shop page 0.369 5.4 s 6.15 s 6.44 s 0%
Category page 0.369 3.56 s 5.33 s 5.64 s 0%
Product page 0.369 3.48 s 5.19 s 5.46 s 0%
Product search 0.369 3.03 s 4.06 s 4.48 s 0%
32 virtual users
Home page 0.426 9.94 s 17 s 17.4 s 0%
Shop page 0.426 18.3 s 21 s 22.7 s 0%
Category page 0.426 18.5 s 21.5 s 22.3 s 0%
Product page 0.426 13.2 s 15.2 s 15.6 s 0%
Product search 0.16 11.4 s 11.7 s 11.7 s 0%

Cart and checkout

Step Requests/s Median p95 p99 Error rate
8 virtual users
Add to cart 0.719 3.95 s 4.63 s 4.96 s 0%
Cart 0.719 3.26 s 4.36 s 4.49 s 0%
Checkout 0.719 2.45 s 3.33 s 3.7 s 0%
32 virtual users
Add to cart 0.869 13.6 s 18.9 s 19.6 s 0%
Cart 0.774 14.1 s 17.7 s 18.6 s 0%
Checkout 0.435 11.2 s 12.3 s 12.3 s 0%

k6 with a fixed number of virtual users, 73 to 75 s per level. The error rate counts responses that weren't a 200.

Discourse

Signed in

Failed: k6 summary at 4 virtual users didn't parse: k6 recorded no requests for step user_activity

Anonymous

Failed: k6 summary at 4 virtual users didn't parse: k6 recorded no requests for step user_activity

Ghost

Readers

Step Requests/s Median p95 p99 Error rate
1 virtual user
Home page 0.863 283 ms 488 ms 573 ms 0%
Post page 2.16 168 ms 385 ms 427 ms 0%
Tag page 0.431 233 ms 476 ms 543 ms 0%
Author page 0.431 214 ms 339 ms 514 ms 0%
RSS feed 0.431 228 ms 359 ms 409 ms 0%
4 virtual users
Home page 1.25 772 ms 1.2 s 1.33 s 0%
Post page 3.13 596 ms 973 ms 1.16 s 0%
Tag page 0.643 480 ms 850 ms 916 ms 0%
Author page 0.643 520 ms 852 ms 1.05 s 0%
RSS feed 0.594 562 ms 821 ms 899 ms 0%
16 virtual users
Home page 1.02 3.83 s 5.08 s 5.98 s 0%
Post page 2.56 3.28 s 4.35 s 4.78 s 0%
Tag page 0.487 2.26 s 3.27 s 3.33 s 0%
Author page 0.503 2.05 s 3.06 s 3.69 s 0%
RSS feed 0.503 2.14 s 2.81 s 3.36 s 0%

Staff in Ghost Admin

Step Requests/s Median p95 p99 Error rate
1 virtual user
Post list 0.456 287 ms 496 ms 585 ms 0%
Post editor 1.4 125 ms 238 ms 300 ms 0%
Saving a draft 1.4 311 ms 501 ms 644 ms 0%
Member list 0.472 323 ms 565 ms 1.02 s 0%
4 virtual users
Post list 0.485 416 ms 734 ms 886 ms 0%
Post editor 1.52 246 ms 502 ms 649 ms 0%
Saving a draft 1.55 1.82 s 2.72 s 3.03 s 0%
Member list 0.518 502 ms 929 ms 963 ms 0%

k6 with a fixed number of virtual users, 60 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 403 MB and MySQL at 599 MB.

Wagtail

Visitors

Step Requests/s Median p95 p99 Error rate
1 virtual user
Home page 0.18 868 ms 1.17 s 1.19 s 0%
Blog index 0.164 2.6 s 3.68 s 3.8 s 0%
Blog post 0.491 192 ms 377 ms 390 ms 0%
Bread page 0.507 236 ms 436 ms 561 ms 0%
Location page 0.164 207 ms 354 ms 373 ms 0%
Site search 0.164 442 ms 858 ms 858 ms 0%
4 virtual users
Home page 0.378 1.4 s 2.17 s 2.47 s 0%
Blog index 0.41 4.74 s 6.07 s 6.3 s 0%
Blog post 1.15 384 ms 720 ms 833 ms 0%
Bread page 1.17 388 ms 780 ms 902 ms 0%
Location page 0.41 448 ms 599 ms 682 ms 0%
Site search 0.378 704 ms 1.69 s 1.83 s 0%
16 virtual users
Home page 0.292 6.17 s 7.02 s 7.55 s 0%
Blog index 0.336 14.6 s 16.6 s 16.8 s 0%
Blog post 0.818 3.33 s 5.59 s 5.96 s 0%
Bread page 0.891 3.46 s 5.8 s 6.19 s 0%
Location page 0.307 3.39 s 5.36 s 5.46 s 0%
Site search 0.263 3.66 s 6.18 s 6.58 s 0%

Editors in the Wagtail admin

Step Requests/s Median p95 p99 Error rate
1 virtual user
Admin dashboard 0.062 813 ms 1.05 s 1.07 s 0%
Page explorer 0.14 3.34 s 4 s 4.06 s 0%
Page editor 0.186 1.33 s 2.46 s 2.84 s 0%
Image library 0.062 638 ms 4.07 s 4.54 s 0%
Admin page search 0.062 374 ms 622 ms 637 ms 0%
4 virtual users
Admin dashboard 0.104 2.4 s 2.65 s 2.7 s 0%
Page explorer 0.239 8.01 s 11.7 s 11.9 s 0%
Page editor 0.343 3.42 s 4.66 s 5.36 s 0%
Image library 0.119 1.39 s 2.45 s 2.74 s 0%
Admin page search 0.104 750 ms 1.22 s 1.29 s 0%

k6 with a fixed number of virtual users, 61 to 68 s per level, after a 25 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 283 MB, Postgres at 174 MB, and Redis at 50 MB.

Advertised vs. measured

Advertised Measured
vCPU 2 2
CPU Shared vCPU 55.5% of busy CPU time stolen
RAM 3 GB 2.9 GiB
Storage 100 GB SSD 96.8 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
2
CPU frequency
2300 MHz
AES-NI
Yes
VM-x/AMD-V
Yes
RAM reported
2.9 GiB
Swap reported
0 MiB
Disk reported
96.8 GiB
Virtualization reported
KVM
CPU steal
55.5% 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
3GB Standard
Product line
Standard VPS
Location
Los Angeles, CA
vCPU
2 shared vCPU
Architecture
x86_64
RAM
3 GB
Storage
100 GB SSD
Virtualization
KVM
Network
Unknown
Transfer
6 TB
Port 25
Unknown
Monthly price at test time
$15.00
Hourly price at test time
$0.022
YABS version
v2026-09-20
Ran at
Oct 17, 2026, 10:55:27 UTC
Parser version
5

Run timeline

Queued
Oct 17, 2026, 10:52:00 UTC
Server created
Oct 17, 2026, 10:53:36 UTC
First update from the server
Oct 17, 2026, 10:53:36 UTC
Completed
Oct 17, 2026, 13:31:00 UTC
Server deleted
Oct 17, 2026, 13:31:00 UTC
Cost of this run
$0.058
TestResultTime
Watch party 0 watching

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

Watch party

Loading the party