Agreed Service Time (AST) is the contractually defined period during which an IT service must be available to users and perform its intended functions. This metric is used to define IT service availability requirements, calculate actual service availability, and monitor compliance with a Service Level Agreement (SLA).
AST may cover continuous service operation (24/7) or be limited to specific hours and days. For example, a corporate information system may have agreed service hours from 9:00 a.m. to 6:00 p.m. on weekdays, while an online store may require 24/7 availability.
What Agreed Service Time Is Used For
AST allows the customer and IT service provider to define the time periods during which service availability commitments apply. This metric is used when establishing SLAs, organizing technical support, planning infrastructure maintenance, and evaluating service quality.
Agreed Service Time serves several purposes:
- Defining availability requirements. Specifies the periods during which the service must operate in accordance with the terms of the agreement.
- Monitoring SLA compliance. Provides the basis for calculating actual availability and identifying breaches of contractual obligations.
- Planning maintenance. Helps schedule preventive maintenance, updates, and other operations in accordance with the agreed service schedule.
- Assessing downtime. Helps determine the duration of service unavailability within the agreed period and its impact on SLA metrics.
How Availability Is Calculated Based on AST
Agreed Service Time is typically measured in hours or minutes over a defined reporting period, such as a week, month, or year.
Actual service availability is calculated using the following formula:
A = (AST − DT) / AST × 100%
where:
- A — actual service availability as a percentage
- AST — Agreed Service Time
- DT — counted downtime within the AST period
For example, a company has an agreement that requires a corporate service to be available 24/7. In a 30-day month, the Agreed Service Time is 720 hours. If an unplanned outage lasting 2 hours occurs during this period, actual availability is:
A = (720 − 2) / 720 × 100% = 99.72%
The specific terms of the SLA must also be taken into account. For example, pre-agreed maintenance windows may be excluded from availability calculations if this is explicitly stipulated in the agreement.
How AST Differs from Actual Operating Time and SLA
These concepts are related but describe different characteristics of an IT service:
- AST (Agreed Service Time) — the agreed period during which the service must be available.
- Uptime — the actual amount of time during which the service was operational.
- Downtime — the duration of service unavailability.
- SLA (Service Level Agreement) — an agreement that defines service quality requirements, including availability, measurement procedures, and the responsibilities of the parties.
For example, an SLA may require availability of at least 99.9% within the Agreed Service Time. If the AST is 720 hours per month, the maximum counted downtime allowed at this availability level is 43 minutes and 12 seconds.
Thus, AST defines the time interval within which compliance with service commitments is assessed, while the availability percentage specified in the SLA determines the allowable duration of downtime.
Where AST Is Used
Agreed Service Time is used in ITSM (IT Service Management), corporate IT infrastructure operations, and cloud service delivery.
The most common use cases include:
- Cloud services and hosting. Defining availability periods for virtual servers, cloud platforms, and hosted applications.
- IT outsourcing. Agreeing on operating hours for supported systems and the obligations of an external service provider.
- Corporate information systems. Establishing availability schedules for ERP, CRM, and other business applications.
- Data centers. Monitoring compliance with contractual availability requirements for hosted IT infrastructure.
Agreed Service Time should not be confused with technical support hours. For example, a service may be available 24/7, while user requests are handled only during business hours.