Simone Casale Brunet

Profiling of Dataflow Programs Using Post Mortem Causation Traces

The natural representation of data streams, parallelism, and composition has made dataflow an attractive programming model for expressing a wide range of stream and media processing applications, and has led MPEG and ISO to base their latest video coding standards on this model. This paper describes and compares methodologies and metrics for the optimization of signal processing algorithms represented as dataflow programs. Our approach is based on the analysis of traces and addresses some of the complexity challenges that arise from the very large data sets that are required for evaluating real-world applications. The methodology and experimental results are demonstrated and evaluated in two at-size case studies, an MPEG-4 SP and an AVC/H.264 video decoders.

Buffer optimization based on critical path analysis of a dataflow program design

The trade-off between throughput and memory constraints is a common design problem in embedded systems, and especially for streaming applications, where the memory in question usually occurs in the form of buffers for streams of data. This paper presents a methodology, based on the post-processing of dataflow execution traces, that enables designers to make principled choices in the design space for arbitrary streaming applications in a scalable manner. It significantly extends the class of applications over traditional compile-time-only techniques, and effectively enables designers to find a close-to-minimum solution for this NP-complete problem. A heuristic algorithm exploring different buffer size configurations lets designers choose appropriate alternatives and enables them to rapidly navigate the design space. Methodology and experimental results are demonstrated in an at-size scenario using a real-world MPEG-4 SP decoder.

Design space exploration of high level stream programs on parallel architectures: A focus on the buffer size minimization and optimization problem

This paper presents a dataflow design methodology and an associated co-exploration environment, focusing on the optimization of buffer sizes. The approach is applicable to dynamic dataflow designs and its performance is presented and validated by experimental results on the porting of an MPEG-4 Simple Profile decoder to the STM STHORM manycore platform. For the purpose of this work, the decoder has been written using the RVC-CAL dataflow language standardized by ISO/IEC. Starting from this high-level representation it is demonstrated how the buffer size configuration can be optimized, based on a novel buffer size minimization algorithm suitable for a very general class of dataflow programs.

Representing Guard Dependencies in Dataflow Execution Traces

Heterogeneous parallel systems are becoming mainstream computing platforms nowadays. One of the main challenges the development community is currently facing is how to fully exploit the available computational power when porting existing programs or developing new ones with available techniques. In this direction, several design space exploration methods have been presented and extensively adopted. However, defining the feasible design space of a dynamic dataflow program still remains an open issue. This paper proposes a novel methodology for defining such a space through a serial execution. Homotopy theoretic methods are used to demonstrate how the design space of a program can be reconstructed from its serial execution trajectory. Moreover, the concept of dependencies graph of a dataflow program defined in the literature is extended with the definition of two new kinds of dependencies - the Guard Enable and Disable - and the 3-tuple notion needed to represent them.

Characterizing communication behavior of dataflow programs using trace analysis

This paper investigates the relationship between two ways of analyzing streaming systems: trace analysis for dataflow programs with firing, and network calculus for network flows. While the former focuses on the structure of the dataflow computation while idealizing the interaction with the environment, the latter characterizes the temporal properties of flows of data, largely abstracting from the computations that connect them. In this work, we show how a trace analysis can be used to derive a description of communication behavior of a dataflow program in terms of network calculus, which is then applied to answering a question about the buffer sizing. The example being studied is an MPEG-4 video decoder.