Measuring availability is subject to a certain degree of interpretation. A system that has been up and running for 365 days in a non-leap year may have been overshadowed by a network outage that lasted 9 hours during a period of peak usage; the user community will see the system as unavailable, while the system administrator will claim 100% “uptime.” However following the true definition of availability, the system will be approximately 99.897% available (8751 hours of the 8760 hours per non-leap year).

Additionally, users may consider a system that is experiencing performance problems to be partially or completely unavailable, even though the system continues to function. Likewise, the non-availability of certain functions of the application may go unnoticed by administrators, although it may be very noticeable to users. Thus, the true measure of availability must be holistic.

Availability must be measured to be determined, ideally with comprehensive monitoring tools ("instrumentation") that in turn present high availability. If there is a lack of instrumentation, systems that are subject to a high volume of transaction processing throughout the day and night (such as credit card processing or telephone switches) are typically inherently better monitored, at least by the same users, than systems that experience periodic lulls in demand.

Recovery time is closely related to availability, which is the total time required for a planned outage or the time required for complete recovery from an unplanned outage. Recovery time can be infinite with certain designs and system failures, full recovery is impossible. An example is a fire or flood that destroys a data center and its systems when there is no secondary data center for disaster recovery.

Another related concept is data availability, which is the degree to which databases and other information storage systems faithfully record and report the transactions of the system. Information management specialists often separately focus on data availability to determine acceptable or actual data loss with various failure events. Some users can tolerate interruptions in application service but not data loss.

Paradoxically, adding more components to the total system can undermine efforts to achieve high availability. This is because complex systems inherently have more potential points of failure and are more difficult to implement correctly. Most highly available systems feature a simple design pattern: a simple, high-quality multipurpose physical system with understandable internal redundancy running all interdependent functions paired with a second system in a separate physical location.

This classic design pattern is common among financial institutions for example. The computing and communications industry has established the Service Availability Forum, which hosts the creation of highly available network infrastructure products, services, and systems. The same basic design principle applies beyond computing in diverse fields such as nuclear power, aeronautics, and healthcare.

The main characteristics of the hierarchical network design are high availability at layer 3 and for its implementation in LAN networks, protocols such as HSRP, GLBP and VRRP are used.

Measuring availability is subject to a certain degree of interpretation. A system that has been up and running for 365 days in a non-leap year may have been overshadowed by a network outage that lasted 9 hours during a period of peak usage; the user community will see the system as unavailable, while the system administrator will claim 100% “uptime.” However following the true definition of availability, the system will be approximately 99.897% available (8751 hours of the 8760 hours per non-leap year).

Additionally, users may consider a system that is experiencing performance problems to be partially or completely unavailable, even though the system continues to function. Likewise, the non-availability of certain functions of the application may go unnoticed by administrators, although it may be very noticeable to users. Thus, the true measure of availability must be holistic.

Availability must be measured to be determined, ideally with comprehensive monitoring tools ("instrumentation") that in turn present high availability. If there is a lack of instrumentation, systems that are subject to a high volume of transaction processing throughout the day and night (such as credit card processing or telephone switches) are typically inherently better monitored, at least by the same users, than systems that experience periodic lulls in demand.

Recovery time is closely related to availability, which is the total time required for a planned outage or the time required for complete recovery from an unplanned outage. Recovery time can be infinite with certain designs and system failures, full recovery is impossible. An example is a fire or flood that destroys a data center and its systems when there is no secondary data center for disaster recovery.

Another related concept is data availability, which is the degree to which databases and other information storage systems faithfully record and report the transactions of the system. Information management specialists often separately focus on data availability to determine acceptable or actual data loss with various failure events. Some users can tolerate interruptions in application service but not data loss.

Paradoxically, adding more components to the total system can undermine efforts to achieve high availability. This is because complex systems inherently have more potential points of failure and are more difficult to implement correctly. Most highly available systems feature a simple design pattern: a simple, high-quality multipurpose physical system with understandable internal redundancy running all interdependent functions paired with a second system in a separate physical location.

This classic design pattern is common among financial institutions for example. The computing and communications industry has established the Service Availability Forum, which hosts the creation of highly available network infrastructure products, services, and systems. The same basic design principle applies beyond computing in diverse fields such as nuclear power, aeronautics, and healthcare.

The main characteristics of the hierarchical network design are high availability at layer 3 and for its implementation in LAN networks, protocols such as HSRP, GLBP and VRRP are used.