Jorn W. Janneck

Taming heterogeneity - the Ptolemy approach

Modern embedded computing systems tend to be heterogeneous in the sense of being composed of subsystems with very different characteristics, which communicate and interact in a variety of ways-synchronous or asynchronous, buffered or unbuffered, etc. Obviously, when designing such systems, a modeling language needs to reflect this heterogeneity. Todays modeling environments usually offer a variant of what we call amorphous heterogeneity to address this problem. This paper argues that modeling systems in this manner leads to unexpected and hard-to-analyze interactions between the communication mechanisms and proposes a more structured approach to heterogeneity, called hierarchical heterogeneity, to solve this problem. It proposes a model structure and semantic framework that support this form of heterogeneity, and discusses the issues arising from heterogeneous component interaction and the desire for component reuse. It introduces the notion of domain polymorphism as a way to address these issues.

Synthesizing hardware from dataflow programs: An MPEG-4 simple profile decoder case study

The MPEG Reconfigurable Video Coding working group is developing a new library-based process for building the reference codecs of future MPEG standards, which is based on dataflow and uses an actor language called CAL. The paper presents a code generator producing RTL targeting FPGAs for CAL, outlines its structure, and demonstrates its performance on an MPEG-4 Simple Profile decoder. The resulting implementation is smaller and faster than a comparable RTL reference design, and the second half of the paper discusses some of the reasons for this counter-intuitive result.

OpenDF: a dataflow toolset for reconfigurable hardware and multicore systems

This paper presents the OpenDF framework and recalls that dataflow programming was once invented to address the problem of parallel computing. We discuss the problems with an imperative style, von Neumann programs, and present what we believe are the advantages of using a dataflow programming model. The CAL actor language is briefly presented and its role in the ISO/MPEG standard is discussed. The Dataflow Interchange Format (DIF) and related tools can be used for analysis of actors and networks, demonstrating the advantages of a dataflow approach. Finally, an overview of a case study implementing an MPEG- 4 decoder is given.

Synthesizing Hardware from Dataflow Programs

The MPEG Reconfigurable Video Coding working group is developing a new library-based process for building the reference codecs of future MPEG standards, which is based on dataflow and uses an actor language called Cal. The paper presents a code generator producing RTL targeting FPGAs for Cal, outlines its structure, and demonstrates its performance on an MPEG-4 Simple Profile decoder. The resulting implementation is smaller and faster than a comparable RTL reference design, and the second half of the paper discusses some of the reasons for this counter-intuitive result.

Actor-oriented control system design: a responsible framework perspective

Complex control systems are heterogeneous, in the sense of discrete computer-based controllers interacting with continuous physical plants, regular data sampling interleaving with irregular communication and user interaction, and multilayer and multimode control laws. This heterogeneity imposes great challenges for control system design in terms of end-to-end control performance modeling and simulation, traceable refinements from algorithms to software/hardware implementation, and component reuse. This paper presents an actor-oriented design methodology that tackles these issues by separating the data-centric computational components (a.k.a. actors) and the control-flow-centric scheduling and activation mechanisms ( a.k.a. frameworks). Semantically different frameworks are composed hierarchically to manage heterogeneous models and achieve actor and framework reuse. We introduce a notion of responsible frameworks to characterize the property that a framework can aggregate individual actors execution into a well-defined composite execution such that heterogeneous models can be composed. This methodology is implemented in the Ptolemy II software environment. We discuss how some of the most useful models for control system design are implemented as responsible frameworks. As an example, the methodology and the Ptolemy II software environment is applied to the design of a distributed, real-time software implementation of a pendulum inversion and stabilization system.

Methods to explore design space for MPEG RMC codec specifications

The recent MPEG Reconfigurable Media Coding (RMC) standard aims at defining media processing specifications (e.g. video codecs) in a form that abstracts from the implementation platform, but at the same time is an appropriate starting point for implementation on specific targets. To this end, the RMC framework has standardized both an asynchronous dataflow model of computation and an associated specification language. Either are providing the formalism and the theoretical foundation for multimedia specifications. Even though these specifications are abstract and platform-independent the new approach of developing implementations from such initial specifications presents obvious advantages over the approaches based on classical sequential specifications. The advantages appear particularly appealing when targeting the current and emerging homogeneous and heterogeneous manycore or multicore processing platforms. These highly parallel computing machines are gradually replacing single-core processors, particularly when the system design aims at reducing power dissipation or at increasing throughput. However, a straightforward mapping of an abstract dataflow specification onto a concurrent and heterogeneous platform does often not produce an efficient result. Before an abstract specification can be translated into an efficient implementation in software and hardware, the dataflow networks need to be partitioned and then mapped to individual processing elements. Moreover, system performance requirements need to be accounted for in the design optimization process. This paper discusses the state of the art of the combinatorial problems that need to be faced at this design space exploration step. Some recent developments and experimental results for image and video coding applications are illustrated. Both well-known and novel heuristics for problems such as mapping, scheduling and buffer minimization are investigated in the specific context of exploring the design space of dataflow program implementations.

Reconfigurable Media Coding: A New Specification Model for Multimedia Coders

Multimedia coding technology, after about 20 years of active research, has delivered a rich variety of different and complex coding algorithms. Selecting an appropriate subset of these algorithms would, in principle, enable a designer to produce the codec supporting any desired functionality as well as any desired trade-off between compression performance and implementation complexity. Currently, interoperability demands that this selection process be hard-wired into the normative descriptions of the codec, or at a lower level, into a predefined number of choices, known as profiles, codified within each standard specification. This paper presents an alternative paradigm for codec deployment that is currently under development by MPEG, known as Reconfigurable Media Coding (RMC). Using the RMC framework, arbitrary combinations of fundamental algorithms may be assembled, without predefined standardization, because everything necessary for specifying the decoding process is delivered alongside the content itself. This side-information consists of a description of the bitstream syntax, as well as a description of the decoder configuration. Decoder configuration information is provided as a description of the interconnections between algorithmic blocks. The approach has been validated by development of an RMC format that matches MPEG-4 Video, and then extending the format by adding new chroma-subsampling patterns.

Automatic software synthesis of dataflow program: An MPEG-4 simple profile decoder case study

The MPEG reconfigurable video coding (RVC) framework is a new standard under development by MPEG that aims at providing a unified high-level specification of current MPEG video coding technologies. In this framework, a decoder is built as a configuration of video coding modules taken from the standard ldquoMPEG toolbox libraryrdquo. The elements of the library are specified by a textual description that expresses the I/O behavior of each module and by a reference software written using the CAL Actor Language. A decoder configuration is written in an XML dialect by connecting a set of CAL modules. Code generators are fundamental supports that enable the direct transformation of a high level specification to efficient hardware and software implementations. This paper presents a synthesis tool that from a CAL dataflow program generates C code and an associated SystemC model. Experimental results of the RVC Expertpsilas MPEG-4 simple profile decoder synthesis are reported. The generated code and the associated SystemC model are validated against the original CAL description which is simulated using the open dataflow environment.

Exploiting statically schedulable regions in dataflow programs

Dataflow descriptions have been used in a wide range of Digital Signal Processing (DSP) applications, such as multi-media processing, and wireless communications. Among various forms of dataflow modeling, Synchronous Dataflow (SDF) is geared towards static scheduling of computational modules, which improves system performance and predictability. However, many DSP applications do not fully conform to the restrictions of SDF modeling. More general dataflow models, such as CAL [1], have been developed to describe dynamically-structured DSP applications. Such generalized models can express dynamically changing functionality, but lose the powerful static scheduling capabilities provided by SDF. This paper focuses on detection of SDF-like regions in dynamic dataflow descriptions — in particular, in the generalized specification framework of CAL. This is an important step for applying static scheduling techniques within a dynamic dataflow framework. Our techniques combine the advantages of different dataflow languages and tools, including CAL [1], DIF [2] and CAL2C [3]. The techniques are demonstrated on the IDCT module of MPEG Reconfigurable Video Coding (RVC).

Profiling dataflow programs

As dataflow descriptions of media processing become popular, the techniques for analyzing and profiling the performance of sequential algorithms are no longer applicable. This paper describes some of the basic concepts and techniques for analyzing the computations described by dataflow programs, and illustrates them on an MPEG-4 decoder.