Distributed computing - JavaTech: An Introduction to Scientific and Technical Computing with Java - page 483

Figure 16.5 Conceptual model showing the model, view, and controller
relationships that are internal to the server architecture.
implements that method must fetch the data from the storage location. Thus, from
the point of view of the data storage component, the front-end server object is
the consumer, or the MVC view. These “view” and “model” relationships are
illustrated in the conceptual model shown in Figure 16.5.
The MVC controller is the component that controls the data. From this point
of view, controlling the model means controlling what is stored there - i.e. in
the data storage component. It is the compute engine that calculates the data
and provides it to the data storage component. The data storage component only
contains data placed there by the compute engine. Thus the compute engine can
be thought of as the controller. In this description, the controller controls only
the data model; it has no direct control over the view component.
16.6.2 Internal server collaborations
The conceptual model developed so far is not complete for at least two reasons.
First, recall that the client must have some way to tell the compute engine what
to compute. The client does so by providing input data via the front-end server
object, which is the only server-side object that the client has access to or even
knows about. Thus there must be some association between the front-end server
and the compute engine. That is, the front-end server must somehow communicate
the input data to the compute engine.
The second shortcoming in the conceptual model is more subtle. Notice that
there must be two threads of execution occurring inside the server. One thread
is the compute engine, calculating data and storing it into the data storage com-
ponent. But any response to the client's request to retrieve data must occur on a