Flexible and adaptive behavior is seen as one of the key characteristics of next generation hard real-time systems. Within the context of fixed priority pre-emptive scheduling, existing approaches deal with optional components and provide kernel mechanisms to schedule effectively such components when spare processor capacity is available. This paper describes a framework that provides a task programming model with optional components, and the appropriate mechanisms for supporting it, by using the main results of existing research in computing spare processor capacity. The paper shows how these ideas can be adapted for being used from an Ada application. The concurrency and real-time programming features of Ada allow an elegant and efficient implementation of a model where hard real-time tasks, optional unbounded components and optional firm tasks coexist.