What is "lazy initialization"

n. (alternative spelling of lazy initialisation English)

In computer programming, lazy initialization is the tactic of delaying the creation of an object, the calculation of a value, or some other expensive process until the first time it is needed.

This is typically accomplished by augmenting a variable's accessor method (or property definition) to check for a previously-created instance. If none exists a new instance is created, placed into the variable, and this new object is returned to the caller in a just-in-time fashion. In this manner object creation is deferred until first use which can, in some circumstances (e.g., sporadic object access), increase system responsiveness and speed startup by bypassing large-scale object pre-allocation. (Note that this may have attendant counter-effects on overall performance, however, as the impact of object instantiation is then amortized across the startup/warm-up phase of the system.)

In multithreaded code, access to lazy-initialized objects/state must be synchronized to guard against a race condition.

See lazy evaluation for a general treatment of this idea. In heavily imperative languages this pattern carries hidden dangers, as does any programming habit that relies on shared state.