---
title: "How to Benchmark Cloud Servers (VPS)"
description: "Run disk, network, and CPU benchmarks on your VPS using YABS and other tools. Compare providers before you commit."
date: 2026-06-25
categories: ["vps"]
tags: ["linux","vps","benchmarks"]
---
import Button from "../../components/widgets/Button.astro";
Before picking a VPS provider, run a benchmark. Specs on a pricing page don't tell you much — shared vCPU contention, noisy neighbors, and disk throttling all affect real performance. A quick benchmark shows what you actually get.
This guide covers what to test, which tools to use, and how to read the results.
## What to benchmark
Three things matter most on a VPS:
**Disk I/O.** Random read/write speeds with small block sizes (4k) affect databases and web apps directly. Sequential speeds (1m blocks) matter for backups and large file operations. Look for NVMe storage — it's standard on most providers now but some still use SATA SSDs.
**Network throughput.** If you're serving users across regions, test speeds to multiple locations. A server with 10 Gbps uplink but poor peering to your target region is useless. Check both IPv4 and IPv6.
**CPU performance.** Geekbench scores give you a standardized number you can compare across providers. Single-core matters for most web workloads. Multi-core matters if you're running parallel tasks.
## YABS: the one-command benchmark
[YABS](https://github.com/masonr/yet-another-bench-script) (Yet-Another-Bench-Script) runs all three tests in one go. It uses fio for disk, iperf3 for network, and Geekbench 6 for CPU. No installation needed — it downloads portable binaries and runs them.
Run it:
```
curl -sL https://yabs.sh | bash
```
That's it. Takes about 10-15 minutes depending on your server.
**Useful flags:**
- `-r` — reduces iperf locations (less bandwidth usage)
- `-i` — skips network tests entirely
- `-f` — skips disk tests
- `-g` — skips Geekbench
- `-5` — runs Geekbench 5 instead of 6 (for comparing with older results)
You can combine flags: `curl -sL https://yabs.sh | bash -s -- -r` runs a lighter test with fewer network endpoints.
### Reading the output
Here's what a typical run looks like (Hetzner CX33, 4 vCPU AMD EPYC, €6.49/month):
```
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
# Yet-Another-Bench-Script #
# v2026-04-20 #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
Thu Oct 16 11:32:39 AM UTC 2026
Basic System Information:
---------------------------------
Uptime : 0 days, 0 hours, 30 minutes
Processor : AMD EPYC-Rome Processor
CPU cores : 4 @ 2445.404 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM : 7.6 GiB
Swap : 0.0 KiB
Disk : 75.0 GiB
Distro : Ubuntu 24.04.3 LTS
Kernel : 6.8.0-71-generic
VM Type : KVM
IPv4/IPv6 : ✔ Online / ✔ Online
fio Disk Speed Tests (Mixed R/W 50/50):
---------------------------------
Block Size | 4k (IOPS) | 64k (IOPS)
------ | --- ---- | ---- ----
Read | 115.01 MB/s (28.7k) | 988.49 MB/s (15.4k)
Write | 115.32 MB/s (28.8k) | 993.69 MB/s (15.5k)
Total | 230.34 MB/s (57.5k) | 1.98 GB/s (30.9k)
| |
Block Size | 512k (IOPS) | 1m (IOPS)
------ | --- ---- | ---- ----
Read | 1.78 GB/s (3.4k) | 2.16 GB/s (2.1k)
Write | 1.88 GB/s (3.6k) | 2.30 GB/s (2.2k)
Total | 3.66 GB/s (7.1k) | 4.46 GB/s (4.3k)
iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider | Location (Link) | Send Speed | Recv Speed | Ping
----- | ----- | ---- | ---- | ----
Clouvider | London, UK (10G) | 5.16 Gbits/sec | 5.60 Gbits/sec | 17.8 ms
Eranium | Amsterdam, NL (100G) | 12.3 Gbits/sec | 12.8 Gbits/sec | 9.27 ms
Uztelecom | Tashkent, UZ (10G) | 1.96 Gbits/sec | 2.24 Gbits/sec | 94.6 ms
Leaseweb | Singapore, SG (10G) | 665 Mbits/sec | 841 Mbits/sec | 166 ms
Clouvider | Los Angeles, CA, US (10G) | 1.03 Gbits/sec | 1.21 Gbits/sec | 158 ms
Leaseweb | NYC, NY, US (10G) | 1.88 Gbits/sec | 2.53 Gbits/sec | 97.7 ms
Edgoo | Sao Paulo, BR (1G) | 616 Mbits/sec | 1.14 Gbits/sec | 219 ms
Geekbench 6 Benchmark Test:
---------------------------------
Test | Value
|
Single Core | 1508
Multi Core | 4919
Full Test | https://browser.geekbench.com/v6/cpu/14484522
YABS completed in 12 min 35 sec
```
**What to look for in these results:**
- **Disk:** 4k random IOPS above 20k is decent for a shared vCPU plan. The 115 MB/s at 4k block size here is typical for Hetzner's cost-optimized tier. Dedicated or regular plans will be higher.
- **Network:** 12+ Gbps to Amsterdam is excellent — that's nearby. The drop to 841 Mbps to Singapore is expected due to distance. Focus on speeds to regions where your users are.
- **CPU:** Geekbench 6 single-core of 1508 is solid for an EPYC-Rome shared vCPU. For reference, dedicated cores typically score 1800-2200+.
## Other benchmark tools worth knowing
YABS covers the basics, but sometimes you need more detail:
**[bench.sh](https://bench.sh/)** — tests disk and network speed. Faster than YABS, no CPU benchmark. Good for a quick check.
```
wget -qO- bench.sh | bash
```
**[nench](https://github.com/n-st/nench)** — similar to bench.sh but adds CPU tests and dual-stack IPv4/IPv6 speed tests by default.
```
curl -sL nench.sh | bash
```
**[fio](https://github.com/axboe/fio)** — the industry standard for disk benchmarking. YABS uses it internally, but running fio directly lets you customize block sizes, queue depths, and test patterns for your specific workload.
```
fio --randrepeat=1 --ioengine=libaio --direct=1 --gtod_reduce=1 \
--name=test --filename=test --bs=4k --iodepth=64 --size=1G \
--readwrite=randrw --rwmixread=75
```
**[Geekbench](https://www.geekbench.com/)** — run it standalone if you only care about CPU scores. Version 6 is current (latest is 6.6 as of February 2026). Scores are standardized and comparable across providers.
**[sysbench](https://github.com/akopytov/sysbench)** — tests CPU, memory, file I/O, and MySQL/PostgreSQL performance. If you're running a database, benchmark it with sysbench before going to production.
## Tips for reliable benchmarks
- **Run tests multiple times.** Shared vCPU performance varies by time of day. Run benchmarks at different hours to see the range.
- **Test during your traffic patterns.** If you'll run production workloads during business hours, benchmark during business hours.
- **Compare same-tier plans.** Don't compare a $4/month shared vCPU against a $40/month dedicated core server. Compare like with like.
- **Check the Geekbench browser.** Search [browser.geekbench.com](https://browser.geekbench.com/) for your provider and plan. Many people post results publicly.
- **Save your results.** YABS outputs a URL you can share. Keep a record so you can compare after provider changes.
## Where to compare providers
These sites collect benchmark results from real users:
- [VPSBenchmarks.com](https://www.vpsbenchmarks.com/) — side-by-side comparisons with real test data
- [VPS Metrics](https://vpsmetrics.com/benchmarks/) — filterable benchmark database
- [LowEndTalk](https://lowendtalk.com/) — community forum where people post YABS results
If you're considering Hetzner specifically, check this [Hetzner Cloud review](https://www.bitdoze.com/hetzner-cloud-review/) with benchmark results across multiple plan tiers.