PTOLEMY II - CiteSeerX

Introduction
rency throughout, allowing for instance for a model to mutate (modify its clustered graph
structure) while the user interface simultaneously modifies the structure in different ways. Consistency
is maintained through the use of monitors and read/write semaphores [59] built upon the
lower level synchronization primitives of Java.
Fully integrated expression language. The Ptolemy II expression language is a higher-order, richly
expressive language that is fully integrated with actor-oriented modeling. The type system inference
mechanism propagates through expressions, parameters, and actor ports seamlessly.
A software architecture based on object modeling. Since Ptolemy Classic was constructed, software
engineering has seen the emergence of sophisticated object modeling [101][123][126] and
design pattern [42] concepts. We have applied these concepts to the design of Ptolemy II, and they
have resulted in a more consistent, cleaner, and more robust design. We have also applied a simplified
software engineering process that includes systematic design and code reviews [119].
1.5.7 Experimental Capabilities
Ptolemy II includes a number of still evolving experimental capabilities. These include.
Distributed models. Ptolemy II has (still preliminary) infrastructure supporting distributed modeling
using CORBA, Java RMI, or lower-level networking primitives. Ptolemy II has (still preliminary)
support for migrating software components.
Higher-order components. Ptolemy II has a (still preliminary) library of actors that operate on
Ptolemy II models rather than just on data. While Ptolemy Classic had such higher-order components,
in Ptolemy Classic the manipulations of the model occurred as part of the initialization
phase of the execution. In Ptolemy II, they can occur during the execution of the model. Examples
include the MobileModel, which supports an actor-oriented version of mobile code, and ModalModel,
which supports modal behavior.
Lifecycle management components. Ptolemy II has a (still preliminary) library of actors that manage
the lifecycle of other Ptolemy II models. Examples include RunCompositeActor, which on
each firing performs a complete execution of the contained model, and ModelReference, which
performs a similar function on a model defined in a separate file or URL.
Component specialization. Ptolemy II has an evolving code generation mechanism that is very different
from that in Ptolemy Classic. In Ptolemy Classic, each component has to have a definition in
the target language, and the code generator merely stitches together these components. In Ptolemy
II, components are defined in Java, and the Java definition is parsed. An API for performing optimization
transformations on the abstract syntax tree is defined, and then compiler back ends can be
used to generate target code. A preliminary implementation of this approach is described in [110],
[134] and [135].
Cal actor definition language. Actors have traditionally been defined in Java for Ptolemy II. However,
static analysis of Java programs for properties that are important at the actor-oriented level is
extremely challenging, at best, and impossible at worst. The Cal actor language provides a way to
define actors so that these properties are statically inferable from the actor definition [35].
Experimental models of computation. Ptolemy II includes a number of experimental models of
computation, including Giotto [52], timed-multitasking [91], distributed discrete-events [31], and
a push-pull component model called component interaction (CI) [141].
38 Ptolemy II