Performance & reliability
Why is my VPS slow?
In order of likelihood: memory exhaustion causing swap thrashing, disk IO saturation, CPU steal time on shared cores, an unoptimised database, or network latency to your users. Check free -h and the st column in top first — between them they identify the cause about eighty per cent of the time.
Memory is the most common culprit and the least obvious. Once the working set exceeds RAM, the kernel swaps, and disk latency replaces memory latency — a factor of ten thousand. The symptom is a server that feels fine and then suddenly does not.
The second most common is disk. Run ioping or fio: NVMe should show sub-millisecond latency. If it does not, either you are on the wrong storage tier or a neighbour is saturating it.
- free -h — if swap is in use and growing, that is the answer
- top — the st column above 2–3% means steal time
- iostat -x 1 — %util near 100 means disk saturation
- Database slow query log — usually a missing index, not a small server
- mtr to a user network — packet loss on a path is not a server problem
See also: CPU steal time · NVMe
7 answers in Performance & reliability
- What is CPU steal time and why does it matter? Steal time is the percentage of time your virtual CPU was ready to run but the hypervisor gave the physical core to someone else. It appears as the "s…
- How much latency should I expect from a VPS? Within a metro, under 5 ms. Within a continent, 10–40 ms. Across the Atlantic, 70–90 ms. Europe to Asia, 150–200 ms. Antipodal routes, 250–320 ms. Lig…
- How many visitors can a VPS handle? A well-configured 2 vCPU / 4 GB instance serves a static site to millions of monthly visits, a cached WordPress site to roughly 300,000, and an uncach…
- How do I benchmark a VPS? Measure the three things that actually vary: single-thread CPU with sysbench, disk with fio at 4K random read/write, and network with iperf3 to a know…
- How much faster is NVMe than SATA SSD? Roughly five times the random IOPS — about 500,000 versus 90,000 — and around five times lower latency, under 0.1 ms versus about 0.5 ms. Sequential t…
- What uptime should I expect from a VPS? Our SLA is 99.99% per instance, which permits about 4.4 minutes of downtime a month, and observed availability across the fleet over the last twelve m…