The complete lifetime of a software system from initial conception through to final obsolescence. The term is most commonly used in contexts where programs are expected to have a fairly long useful life, rather than in situations such as experimental programming where programs tend to be run a few times and then discarded. Traditionally the life cycle has been modelled as a number of successive phases, typically:

• user requirements

• system requirements

• software requirements

• overall design

• detailed design

• component production

• component testing

• integration and system testing

• acceptance testing and release

• operation and maintenance

Such a breakdown tends to obscure several important aspects of software production, notably the inevitable need for iteration around the various life-cycle activities in order to correct errors, modify decisions that prove to have been misguided, or reflect changes in the overall requirements for the system. It is also somewhat confusing to treat operation and maintenance as just another life-cycle phase since during this period it may be necessary to repeat any or all of the activities required for initial development of the system. There has therefore been a gradual movement toward more sophisticated models of the software life cycle. These provide explicit recognition of iteration, and often treat the activities of the operation and maintenance period simply as iterations occurring after rather than before release of the system for operational use. See also spiral model, V-model, waterfall model.