Server Downtime

What Does It Mean?

Server downtime tracks the amount of time your organizations are offline for any reason. Server outages can both be planned (due to maintenance, updates, resets, and other IT-related tasks) or unexpected (due to connectivity issues, system crashes, and other problems).

ServiceNow_Dashboard_Screenshot
A ServiceNow Weekly Analysis Dashboard

Why Does it Matter?

For most businesses and organizations today, being connected and online means having access to key files, services, and tools that are vital for operations. If you offer a web-based service, being offline might mean you’re actively losing money and potentially even clients. Even when talking about internal servers, outages cause serious disruptions. It’s also an important metric potential your service level agreements, as uptime is usually guaranteed at a certain rate. For most purposes, the desired minimum uptime is around 99.9%, with anything below that considered significant.

How Do you Measure the KPI?

To measure the KPI, you can track server downtime either as a comprehensive figure (including both planned and unplanned outages), or measure each individually. In the former case, simply add up all the times your servers were offline for the desired measurement period (daily, weekly, monthly, or yearly). You can build a server downtime monitor into your IT operations analytics tool and track it both as a percentage (uptime) or in minutes (time offline).

What Sources Would You Use to Measure the KPI?

The metrics required to track your server downtime are all performance-related. The first source to include in your server’s log, which will have a record of all connections made, as well as any time it was unable to connect. Additionally, you can collect incident reports related to the server, as well as IT data related to planned outages, along with projections and actual times. Finally, you should always keep logs of unplanned outages, and include them.

Give me an example…

Let’s say you offer a SaaS office communication suite, and your SLA states that you’ll provide 99.95% uptime guaranteed or face financial penalties for extended outages. The 99.95% includes the estimates for your planned outages (system updates, adding new features, server maintenance, and more). Keeping your clients and providing a service that works relies on you being able to keep your servers online at your SLA level or better. Measuring your server downtime is vital to find areas for improvement and avoid missing your SLA.

What Benchmarks/Indicators Should I Use?

- Planned outages - Unplanned downtime - Total time to repair


More IT KPIs Start Free Trial