Service Level Agreement Calculator: Measure SLA Uptime

A service level agreement (SLA) calculator works out the percentage of time a service was available from its total measurement period and recorded downtime. Enter the period and downtime in matching units; the calculator returns the resulting availability percentage, along with the unavailable time and available time.

Formula

Availability (%) = 100 × (measurement period − downtime) ÷ measurement period. Here, the measurement period is the total interval being assessed, in hours; downtime is the part of that interval counted as unavailable, also in hours. The result is a percentage. The related downtime share is 100 × downtime ÷ measurement period, and available time is measurement period minus downtime.

The formula assumes the downtime is within the period and that both values use the same time unit. The calculator requires nonnegative inputs but does not enforce downtime being less than or equal to the period. If downtime exceeds the period, the result is not meaningful; check the inputs before relying on it.

Worked Example

Using the calculator’s defaults, the measurement period is 720 hours and recorded downtime is 7.2 hours.

1. Subtract downtime from the period: 720 − 7.2 = 712.8 hours available.
2. Divide available time by the measurement period: 712.8 ÷ 720 = 0.99.
3. Convert the fraction to a percentage: 0.99 × 100 = 99.00% availability.

The downtime share is 7.2 ÷ 720 × 100 = 1.00%, and available time is 712.80 hours. The availability and downtime shares total 100% when downtime is within the period.

What an SLA Availability Result Means

An SLA is an agreement that describes service expectations and how performance is measured. Availability is one commonly used metric, but the calculator measures only the simple proportion of a chosen period that was not counted as downtime. It does not determine whether a provider met an agreement, whether an outage qualifies under its terms, or whether a remedy is owed.

For example, a 720-hour period is 30 days of 24 hours each. If 7.2 hours count as downtime, the arithmetic gives 99.00% availability. That result is useful only if the agreement uses that same 720-hour window and treats those 7.2 hours as downtime. A different definition or measurement window can produce a different SLA result from the same service history.

Availability is often expressed as “nines,” such as 99.9%, but a percentage alone can obscure the time involved. At 99.9% availability over 30 days of 720 hours, the corresponding downtime allowance is 0.1% × 720 = 0.72 hours, or 43.2 minutes. This is a mathematical illustration, not a claim about any particular contract.

How to Choose the Measurement Period

Use the interval named in the agreement or the one relevant to your analysis. Common windows include a calendar month, a rolling period, a quarter, or a year. Do not substitute a convenient period if the contract specifies another one. For a calendar month, the number of hours varies: a 28-day month has 672 hours, a 30-day month has 720, and a 31-day month has 744. A period that includes a daylight-saving clock change may also need careful handling if records are stored as local times rather than elapsed time.

• Confirm the start and end timestamps and the time zone used for both.
• Use elapsed time consistently. If downtime is recorded in minutes, either convert it to hours or calculate with the whole period in minutes.
• Check whether the agreement uses calendar time, scheduled service hours, or another defined window.
• Keep the underlying incident records so the downtime total can be traced and reviewed.

The calculator’s default of 720 hours is a convenient example for a 30-day period, not a universal monthly value. Replace it with the actual period length for the interval you are measuring.

What Counts as Downtime?

There is no single definition that applies to every SLA. Contract terms may specify how service availability is detected, what constitutes an outage, how partial failures are treated, and whether planned maintenance or customer-caused interruptions are excluded. The calculation cannot interpret these terms. Apply the agreement’s definition consistently before adding incident durations.

Some services can be partly impaired rather than fully unavailable. A simple uptime percentage generally treats each moment as either available or down. If the contract defines service levels by affected users, requests, regions, or severity, a single downtime total may not represent the agreed measure. Review the documented measurement method rather than assuming that every reported incident counts equally.

Be careful not to count overlapping incidents twice. If two outages happen at the same time and both affect the same service measurement, their combined duration may be shorter than the sum of their individual durations. Whether to combine them depends on the agreed method, but the raw elapsed time should be reconciled before entering a total.

Common Calculation Mistakes and Limits

• Mixing units: Comparing 720 hours with 30 minutes produces a wrong result unless minutes are converted to hours (30 minutes = 0.5 hours).
• Using the wrong denominator: A month, a 30-day period, and scheduled business hours are not interchangeable.
• Rounding too early: Add incident durations at a useful precision, calculate the percentage, and round the final display. The calculator displays two decimal places, which can hide small differences.
• Entering a percentage as a fraction: Enter downtime as a duration, not as “1” to mean 1%. The formula converts the duration into a percentage.
• Assuming the result proves compliance: The arithmetic does not account for exclusions, service credits, reporting rules, or other contract terms.

This calculator also treats all hours in the period equally. It cannot calculate request success rates, response-time targets, tiered availability, or weighted service impact unless those measures are separately defined and calculated. It is a transparent arithmetic aid, not a substitute for the written SLA or a complete monitoring report.

Frequently Asked Questions

How do I calculate SLA availability?
Subtract eligible downtime from the measurement period, divide by the full period, and multiply by 100. For example, 7.2 hours down in a 720-hour period gives 99.00% availability.

How much downtime does 99.9% availability allow?
Multiply the period by 0.001. Over 720 hours, that is 0.72 hours, or 43.2 minutes. The allowance changes with the period length.

Should planned maintenance count as downtime?
Check the SLA’s definition. Some agreements exclude specified maintenance; others may count it. The calculator cannot decide whether an event qualifies.

Can downtime be entered in minutes?
Yes, if the measurement period is also expressed in minutes. Alternatively, convert the downtime to hours before using a period entered in hours.

Does 99% availability mean the SLA was met?
Not necessarily. It means the basic availability formula returned 99% for the inputs. Whether that meets the agreement depends on its target, measurement rules, exclusions, and reporting period.

Related

• Advanced Reach Calculator: What It Measures
• Unit Rate Calculator With Answers and Examples
• SLA Availability Calculator: Uptime and Downtime

in period | Measurement period | hours | 720 | 0
in down | Recorded downtime | hours | 7.2 | 0
out availability | Availability | % | 100 * (period - down) / period | 2
out downtime_pct | Downtime share | % | 100 * down / period | 2
out available | Available time | hours | period - down | 2