Skip to main content

> Term

Uptime

The percentage of time a system is available and successfully serving its intended function.

Detailed Explanation

Uptime is the foundational metric for system reliability, usually expressed as a percentage over a specific period, such as a month or a year. While it measures basic availability, modern engineering also evaluates the quality of that uptime through SLIs and SLOs to ensure the system is not just 'up' but actually usable.

Why It Matters

It directly impacts user trust and revenue; frequent or prolonged downtime can lead to customer churn and breaches of Service Level Agreements (SLAs).

Common Failure Mode

Focusing solely on server uptime (e.g., 'the machine is running') while ignoring application-level failures (e.g., 'the login button is broken'), resulting in the 'watermelon effect'—green on the dashboard, red for the user.

Practical Example

A payment gateway reports 99.99% uptime for the month, meaning it experienced roughly 4.3 minutes of total unavailability, successfully meeting its enterprise SLA requirements.

Production Manifestation

Tracked via external synthetics and internal observability tools, often displayed on public status pages to communicate reliability to customers.

Frequently Asked Questions

What is Uptime in short?

The percentage of time a system is available and successfully serving its intended function.

What is the most common failure mode?

Focusing solely on server uptime (e.g., 'the machine is running') while ignoring application-level failures (e.g., 'the login button is broken'), resulting in the 'watermelon effect'—green on the dashboard, red for the user.

AI Summary

The percentage of time a system is available and successfully serving its intended function. It directly impacts user trust and revenue; frequent or prolonged downtime can lead to customer churn and breaches of Service Level Agreements (SLAs).