{"id":401,"date":"2026-06-11T04:06:10","date_gmt":"2026-06-11T04:06:10","guid":{"rendered":"https:\/\/virtualserversvps.com\/blog\/vps-performance-benchmarks-cpu-ram-disk-io\/"},"modified":"2026-09-15T22:03:11","modified_gmt":"2026-09-15T22:03:11","slug":"vps-performance-benchmarks-how-to-test-cpu-ram-and-disk-io","status":"publish","type":"post","link":"https:\/\/virtualserversvps.com\/blog\/vps-performance-benchmarks-how-to-test-cpu-ram-and-disk-io\/","title":{"rendered":"How to Benchmark a VPS Properly: CPU, Memory, Disk IOPS, and Network in 20 Minutes"},"content":{"rendered":"\n

“4 vCPU \/ 8 GB RAM \/ NVMe” tells you the shape of a VPS but nothing about how fast it will actually run. Two hosts selling identical spec sheets can differ by 40% in real throughput because of hypervisor overhead, CPU steal from noisy neighbours, storage contention, and network path quality. The only way to know what you bought is to measure it.<\/p>\n\n\n\n

This is a complete, repeatable benchmark suite you can run in about 20 minutes on a fresh VPS. Run it before you commit to a long contract, and again after deployment so you have a baseline to compare against when something “feels slow” later.<\/p>\n\n\n\n

Before You Start<\/h2>\n\n\n\n
sudo apt update && sudo apt install -y sysbench fio iperf3 htop curl jq bc<\/code><\/pre>\n\n\n\n
Run benchmarks on an idle system, not while your application is building caches or running backups. And run each test at least twice \u2014 the second result is usually the honest one, once the caches and CPU have warmed up.<\/p>\n\n\n\n
1. CPU: Throughput and Steal Time<\/h2>\n\n\n\n
sysbench<\/code> measures raw compute in events per second. Prime numbers are the standard workload because they are deterministic and comparable across machines:<\/p>\n\n\n\n
# Single-core\nsysbench cpu --cpu-max-prime=20000 --threads=1 run<\/p>\n\n\n\n
# All cores\nsysbench cpu --cpu-max-prime=20000 --threads=$(nproc) run<\/code><\/pre><\/p>\n\n\n\n
Compare the multi-core score to the single-core score multiplied by core count. If multicore scaling is poor, the provider may be over-committing physical cores.<\/p>\n\n\n\n
Then check steal time \u2014 the percentage of CPU time your VM wanted but the hypervisor gave to someone else:<\/p>\n\n\n\n
top -bn1 | head -5\n# or\nmpstat 1 5<\/code><\/pre>\n\n\n\n
Under 1% steal is healthy. Sustained 5\u201310% means noisy neighbours and explains erratic latency. Above 10%, open a support ticket with your measurements; that is not a configuration problem you can fix.<\/p>\n\n\n\n
2. Memory: Bandwidth and Latency<\/h2>\n\n\n\n
Throughput per core tells you how much memory bandwidth the host actually exposes to your VM:<\/p>\n\n\n\n
sysbench memory --memory-block-size=1M --memory-total-size=10G \\\n  --memory-oper=read --threads=$(nproc) run<\/p>\n\n\n\n
sysbench memory --memory-block-size=1M --memory-total-size=10G \\\n  --memory-oper=write --threads=$(nproc) run<\/code><\/pre><\/p>\n\n\n\n
Watch the “MiB\/sec” figure. A modern host should deliver tens of GB\/s aggregate; anything far below that on a multi-core plan suggests memory channel throttling or a busy host.<\/p>\n\n\n\n
3. Disk: IOPS, Latency, and Throttling<\/h2>\n\n\n\n
Disk is where VPS performance usually falls apart, because storage is the most heavily shared resource. fio<\/code> gives you the real numbers \u2014 random reads, sequential writes, and mixed workloads:<\/p>\n\n\n\n
# Random 4K read IOPS (typical database pattern)\nfio --name=randread --ioengine=libaio --iodepth=32 --rw=randread \\\n    --bs=4k --direct=1 --size=1G --numjobs=4 --runtime=30 --group_reporting<\/p>\n\n\n\n
# Sequential write throughput (backups, logs)\nfio --name=seqwrite --ioengine=libaio --iodepth=16 --rw=write \\\n    --bs=1M --direct=1 --size=2G --runtime=30 --group_reporting<\/p>\n\n\n\n
# Mixed 70\/30 read\/write (web application pattern)\nfio --name=mixed --ioengine=libaio --iodepth=32 --rw=randrw --rwmixread=70 \\\n    --bs=4k --direct=1 --size=1G --runtime=30 --group_reporting<\/code><\/pre><\/p>\n\n\n\n
Read these fields in the output:<\/p>\n\n\n\n