Actor Model

The Actor Model was originally proposed by Carl Hewitt et. al. in the 70es and since then has evolved into different families ^[1].

We focus here on the classical Actor Model described by Gul Agha in Actors ^[2]. This sees an actor as the universal primitive of concurrent computation:

An actor is a computational entity that, in response to a message it receives, can concurrently:

• send a finite number of messages to other actors;
• create a finite number of new actors;
• designate the behavior to be used for the next message it receives. ^[3]

Actor Characteristics

Actor systems and actors have the following basic characteristics^[4] :

• communication via direct asynchronous messaging,
• actors support finite state machines,
• actors do not share their mutable state,
• lock-free concurrency,
• actors support parallelism,
• actors tend to come in systems.

Modern actor implementations extend those by

• location transparency: this enables actors of different kinds and on different systems to communicate,
• supervision: actors form a dependency relationship where a parent actor supervises child/subordinate ones. This is used to implement fault tolerant systems.
• actor protocols: there are predefined actors following a specific message protocol to implement complex behaviors e.g. as generic server, supervisor, router, publish-subscribe ...

Actor Primitives

The actor machinery is based on only a few basic primitives defined in ActorInterfaces.Classic:

A user can write actor programs using only those. His programs should run with minor modifications on other actor libraries supporting the same basic interface.