The VPS performance numbers behind the price.
Every figure below comes from our own fleet telemetry, sampled hourly and averaged over the last 90 days. The raw exports are on the status page.
- Single-thread throughput
- 2,240
- Random read IOPS
- 512,000
- CPU steal
- 1.4 %
- Uptime, 24 months
- 99.987 %
Single-thread throughput
Single-thread throughput
Geekbench 6 single-core, dedicated-core instances, median of the fleet
Geekbench 6 — single core
Geekbench 6 — multi core, by plan
Storage
Storage
fio 4K random read, queue depth 32, on a loaded host
4K read latency
0.08 ms
Sequential read
6.8 GB/s
Sequential write
4.1 GB/s
Drive endurance
1.3 DWPD
CPU steal
CPU steal
Fleet-wide 95th percentile. We cap host oversubscription at 3:1 on shared plans and 1:1 on dedicated-core plans.
- cpu
- AMD EPYC 9354P — Genoa, Zen 4, 3.25 GHz base / 3.80 GHz boost
- memory
- 384 GB DDR5-4800 ECC registered per host
- storage
- Samsung PM9A3 Gen4 NVMe, RAID-10, hot spare per chassis
- network
- Dual 25 GbE to separate leaf switches, LACP
- power
- A+B feeds, N+1 UPS, on-site diesel with 72 h fuel
- hypervisor
- KVM on Linux 6.12 LTS, nested virtualisation enabled
- oversubscription
- 3:1 on fair-share plans, 1:1 on dedicated-core plans
Run your own benchmark
Run your own benchmark
Deploy anything, run whatever suite you trust, and if the numbers disappoint you, ask for your money back within 72 hours. No form, no reason required.
01How can it be this cheap and still be fast?
02Fair-share or dedicated cores — which do I need?
03What CPU steal should I expect?
04Are backups included?
05Can I resize a server mid-term?
06Do you offer Windows?
Benchmarks are honest and therefore boring: they were run on production hosts carrying real customer load, not on an empty box at 3 a.m.