Home > SOA Glossary > Runtime Autonomy

Runtime Autonomy

The level of control a service has over its processing logic at the time the service is invoked and executing is called runtime autonomy.

The Service Autonomy design principle advocates increasing runtime autonomy in order to guarantee the following to service consumers:

  • consistently acceptable runtime execution performance
  • a greater degree of performance reliability
  • the option for it to be isolated in response to specific security, reliability, or performance requirements
  • a greater level of behavioral predictability (especially when concurrently accessed)

The more a service is comprised of logic or resources that are shared by other parts of the enterprise, the less it is able to make the types of quality of service claims just listed.

The primary reason these claims are so important to service-orientation is service composition. Because a composition exists as an aggregate of programs (services) that may also be participating in other compositions, it tends to be naturally non-autonomous.

See also: