Uptime
Uptime is the percentage of time a website or online service stays accessible and responds correctly, measured over a period such as a day, month, or year. The opposite term, downtime, tracks how long a site was unreachable or failing during that same period. Because uptime is usually expressed as a percentage — for example, "99.9% uptime over the past year" — that percentage can be converted directly into an actual amount of downtime, which is often more useful for judging how serious a given number really is.
How uptime is measured
Uptime is calculated as (time the service was reachable and working ÷ total time in the period) × 100. On its own, that formula only tells you what happened after the fact — the value of continuous monitoring is catching the down periods as they happen, not reconstructing them afterward. Monitoring tools like HostTracker run checks at a fixed interval from many locations around the world, so a single request failing from one region does not immediately count as an outage; the check is repeated and confirmed from other locations before an incident is recorded, which keeps the reported uptime figure accurate instead of inflated by one-off network blips.
What good uptime looks like
Uptime percentages are often described in "nines" — 99.9% is "three nines," 99.99% is "four nines." Converting common tiers into actual downtime, based on a 365-day year:
- 90% uptime = 36.5 days of downtime per year
- 99% uptime ("two nines") = 3 days 15.6 hours of downtime per year
- 99.5% uptime = 1.83 days of downtime per year
- 99.9% uptime ("three nines") = 8 hours 45 minutes 36 seconds of downtime per year
- 99.95% uptime = 4 hours 22 minutes 48 seconds of downtime per year
- 99.99% uptime ("four nines") = 52 minutes 34 seconds of downtime per year
- 99.999% uptime ("five nines") = 5 minutes 15 seconds of downtime per year
Even 99% uptime, which sounds close to perfect, still adds up to several days of outage a year — enough to lose real customers if it happens in one continuous stretch rather than scattered seconds.
The gap between an SLA and monitored uptime
A hosting provider's advertised uptime, such as a "99.9% SLA," is a claim made by the host about its own infrastructure. Monitored uptime is different: it is measured independently, by external checks that run from outside the host's network, at a fixed interval, from multiple global locations. That distinction matters because a host has no particular incentive to report its own failures quickly, or at all, while an independent monitor catches the outage the moment a check fails, whether or not the provider's own status page ever mentions it.
Uptime is closely related to, but not identical to, availability: availability describes whether a system is reachable right now, while uptime is the historical percentage over a period of time.
How HostTracker checks uptime
HostTracker runs uptime checks from 300+ monitoring locations worldwide, using any of 14 check types — HTTP/S, ping, TCP port, transaction, and more — depending on what "working" means for your service. Run a free HTTP check to see current status, or set up continuous distributed monitoring so every outage is caught and logged automatically. For the background on why external monitoring matters, see our guide on why websites need monitoring.