The parallel programming model is
essentially based on functional principles which is sufficient for many kinds
of computer algebra algorithms:
dist[start](f, a, ...)
creates a task evaluating
the expression f(a, ...) and returns a reference t to this
task. The execution of f may take place on any machine connected to the
distributed session and should therefore not cause any side effects. Tasks may
create other tasks and arbitrary Maple objects (including task references) may
be passed as task arguments and returned as task results.
dist[wait](t)
blocks the execution of the current task
until the task represented by t has terminated and returns its
result. Multiple tasks may independently wait for and
retrieve the result of the same task t.
When and on which machine a task is scheduled for execution is entirely in the
responsibility of the underlying runtime system.
Maintained by: Wolfgang Schreiner Last Modification: April 22, 1999