Citation Context

...n parallel with the object's internal process. To control access to variables shared by different operations, the shared variables and the operations manipulating them can be defined within a monitor =-=[10, 16]-=-. Processes synchronize through builtin condition objects. An object's process executes outside of the monitor, but can invoke monitored operations should it need access to shared state. Each object h...

Citation Context

...et of (ρ, [v0, . . . , vn]) such that E evaluates to the value list [v0, . . . , vn] along ρ. We assume that the semantic functions are given, by structural induction, in the usual way. For example: [=-=[10]-=-] = {(δ, 10)} [[i]] = {(i=v, v) | v ∈ Vint} [[e1 + e2]] = {(ρ1ρ2, v1 + v2) | (ρ1, v1) ∈ [[e1]] & (ρ2, v2) ∈ [[e2]]} [[(e0, . . . , en)]] = {(ρ0 . . . ρn, [v0, . . . , vn]) | ∀j. 0 ≤ j ≤ n ⇒ (ρj, vj) ∈...

Citation Context

...venting dangerous concurrent access, the early attempts to embed monitors into programming languages were applied to very static and restricted languages. In particular, monitors in Concurrent Pascal =-=[3]-=- were restricted to passing scalars and arrays of scalars, and programs were statically loaded, statically instantiated. While such a language was sufficient to build a small-scale demonstration of an...

Citation Context

...nment. (4) A real-time program never terminates but continues to serve its environment as long as the computer works. (The occasional need to stop a real-time program, say at the end of an experiment, can be handled by ad hoc mechanisms, such as turning the machine off or loading another program into it.) What is needed then for real-time applications is the ability to specify a fixed number of concurrent tasks that can respond fast to nondeterministic requests. The programming languages Concurrent Pascal and Modula come close to satisfying the requirements for abstract concurrent programming [1, 2, 10]. Both of them are based on the monitor concept [3, 7]. Modula, however, is primarily oriented towards multiprogramming on a single processor. And a straightforward implementation of Concurrent Pascal requires a single processor or a multiprocessor with a common store. In their present form, these languages are not ideal for a microcomputer network with distributed storage only. It may well be possible to modify Concurrent Pascal to satisfy the constraints of distributed storage. The ideas proposed here are more attractive, however, because they unify the monitor and process concepts and resul...