Understanding Uptime Percentage

Understanding Uptime Percentage

Uptime percentage is the most commonly used metric for describing service reliability. This article explains how PulseAPI calculates it and how to interpret the numbers.


How Uptime Percentage Is Calculated

Uptime percentage = (successful checks ÷ total checks) × 100, over a specified time period.

A check is successful if it:

  1. Receives a response from the endpoint
  2. The response's HTTP status code matches the Expected Status Code configured on the monitor
  3. The response is received within the configured Timeout

Any check that times out, returns the wrong status code, or encounters a connection error is counted as a failure.

Example:

  • 1,440 checks in 24 hours (one every 60 seconds)
  • 1,432 successful, 8 failed
  • Uptime = (1,432 ÷ 1,440) × 100 = 99.44%

Time Periods

The uptime percentage displayed on monitor detail pages is calculated for different time periods:

  • 24 hours — reflects recent performance
  • 7 days — week-over-week view
  • 30 days — monthly reliability metric

You can switch between time periods using the selector on the monitor detail page.


What Uptime Numbers Mean in Practice

Uptime % Max Downtime per Year Notes
99% ~3.65 days Often called "two nines"
99.5% ~1.83 days
99.9% ~8.77 hours Common SLA target
99.95% ~4.38 hours
99.99% ~52.6 minutes "Four nines" — common for critical services
99.999% ~5.26 minutes "Five nines" — requires very high availability infrastructure

Impact of Check Interval

Faster check intervals give you more data points, which means:

  • More accurate uptime calculation (fewer "gaps" in data)
  • Faster detection of downtime (and therefore shorter outages affect the percentage less)

With a 60-second interval, a 2-minute outage represents 2 failed checks out of however many total. With a 5-second interval, the same outage represents 24 failed checks — a more accurate reflection of the actual downtime.


Uptime and Alert Rules

You can create an alert rule that fires when uptime drops below a threshold over a time window (e.g., "alert if uptime falls below 99.9% over the last 24 hours"). This is useful for SLA-based alerting rather than per-check alerting.

See Alert Rule Conditions: Uptime Percentage.


Related Articles


Still have questions? Contact support.

    • Related Articles

    • Response Time Chart and Uptime Heatmap

      The monitor detail page includes two visualizations: a Response Time Chart and an Uptime Heatmap. This article explains how to read them and switch between time periods. Response Time Chart The response time chart shows how long checks took over ...
    • Alert Rule Conditions: Uptime Percentage

      An uptime percentage alert rule fires when your monitor's uptime over a specified time window drops below a threshold. This is ideal for SLA-based alerting and reducing false positives from one-off check failures. How It Works Instead of evaluating ...
    • Check Intervals Explained

      The check interval is how often PulseAPI sends a request to your monitor. This article explains what intervals are available, how your plan affects them, and how to choose the right interval for each monitor. What Is a Check Interval? Every monitor ...
    • Viewing Check History

      The check history is a log of every HTTP request PulseAPI has made to a monitor. It's the primary tool for diagnosing problems, understanding patterns, and reviewing past incidents. How to Access Check History In the left sidebar, click Monitors. ...
    • Understanding Response Times

      Every check PulseAPI performs records how long the HTTP request took. This article explains what response time means, what affects it, and how to interpret the data. What PulseAPI Measures Response time is measured from the moment PulseAPI initiates ...