Christian Pilato

Ant Colony Heuristic for Mapping and Scheduling Tasks and Communications on Heterogeneous Embedded Systems

To exploit the power of modern heterogeneous multiprocessor embedded platforms on partitioned applications, the designer usually needs to efficiently map and schedule all the tasks and the communications of the application, respecting the constraints imposed by the target architecture. Since the problem is heavily constrained, common methods used to explore such design space usually fail, obtaining low-quality solutions. In this paper, we propose an ant colony optimization (ACO) heuristic that, given a model of the target architecture and the application, efficiently executes both scheduling and mapping to optimize the application performance. We compare our approach with several other heuristics, including simulated annealing, tabu search, and genetic algorithms, on the performance to reach the optimum value and on the potential to explore the design space. We show that our approach obtains better results than other heuristics by at least 16% on average, despite an overhead in execution time. Finally, we validate the approach by scheduling and mapping a JPEG encoder on a realistic target architecture.

A Survey and Evaluation of FPGA High-Level Synthesis Tools

High-level synthesis (HLS) is increasingly popular for the design of high-performance and energy-efficient heterogeneous systems, shortening time-to-market and addressing todays system complexity. HLS allows designers to work at a higher-level of abstraction by using a software program to specify the hardware functionality. Additionally, HLS is particularly interesting for designing field-programmable gate array circuits, where hardware implementations can be easily refined and replaced in the target device. Recent years have seen much activity in the HLS research community, with a plethora of HLS tool offerings, from both industry and academia. All these tools may have different input languages, perform different internal optimizations, and produce results of different quality, even for the very same input description. Hence, it is challenging to compare their performance and understand which is the best for the hardware to be implemented. We present a comprehensive analysis of recent HLS tools, as well as overview the areas of active interest in the HLS research community. We also present a first-published methodology to evaluate different HLS tools. We use our methodology to compare one commercial and three academic tools on a common set of C benchmarks, aiming at performing an in-depth evaluation in terms of performance and the use of resources.

Ant colony optimization for mapping and scheduling in heterogeneous multiprocessor systems

Heterogeneous multiprocessor systems, assembled with off-the-shelf processors and augmented with reprogrammable devices, thanks to their performance, cost effectiveness and flexibility, have become a standard platform for embedded systems. To fully exploit the computational power offered by these systems, great care should be taken when deciding on which processing element (mapping) and when (scheduling) executing the program tasks. Unfortunately, both these problems are NP-complete, and, even if they are strictly interconnected, they are normally performed separately with exact or heuristic algorithms to simplify the search for the optimum points. In this paper we present an exploration algorithm based on Ant Colony Optimization (ACO) that tries to solve the two problems simultaneously. We propose an implementation of the algorithm that gradually constructs feasible solution instances and searches around them rather than exploring a structure that already considers all the possible solutions. We introduce a two-stage decision mechanism that simplifies the data structures but lets the ant perform correlated choices for both the mapping and the scheduling. We show that this algorithm provides better and more robust solutions in less time than the Simulated Annealing and the Tabu Search algorithms, extended to support the combined scheduling and mapping problems. In particular, our ACO formulation can find, on average, solutions between 64% and 55% better than Simulated Annealing and Tabu Search.

HArtes: Hardware-Software Codesign for Heterogeneous Multicore Platforms

Developing heterogeneous multicore platforms requires choosing the best hardware configuration for mapping the application, and modifying that application so that different parts execute on the most appropriate hardware component. The hArtes toolchain provides the option of automatic or semi-automatic support for this mapping. During test and validation on several computation-intensive applications, hArtes achieved substantial speedups and drastically reduced development times.

Mapping and scheduling of parallel C applications with ant colony optimization onto heterogeneous reconfigurable MPSoCs

Efficient mapping and scheduling of partitioned applications are crucial to improve the performance on todays reconfigurable multiprocessor systems-on-chip (MPSoCs) platforms. Most of existing heuristics adopt the Directed Acyclic (task) Graph as representation, that unfortunately, is not able to represent typical embedded applications (e.g., real-time and loop-partitioned). In this paper we propose a novel approach, based on Ant Colony Optimization, that explores different alternative designs to determine an efficient hardware-software partitioning, to decide the task allocation and to establish the execution order of the tasks, dealing with different design constraints imposed by a reconfigurable heterogeneous MPSoC. Moreover, it can be applied to any parallel C application, represented through Hierarchical Task Graphs. We show that our methodology, addressing a realistic target architecture, outperforms existing approaches on a representative set of embedded applications.

A Multi-objective Genetic Algorithm for Design Space Exploration in High-Level Synthesis

This paper presents a methodology for design space exploration (DSE) in high-level synthesis (HLS), based on a multi-objective genetic algorithm. Since all high-level synthesis sub-tasks are notoriously NP-complete and interdependent and the design objectives are in conflict for nature, most of the already proposed approaches are not efficient in the exploration of this design space and not effective in the identification of different trade-offs. For these reasons, evolutionary algorithms can be considered as good candidates to tackle such difficult explorations. Therefore, we will compare our proposed approach, using different solution encoding, with a publicly available HLS framework and we will show that this approach is able to obtain better optimization results, with respect to the design objectives (latency and area have been considered for optimization), in most of situations and our proposed encoding better approaches the situations when multi-modal functional units (e.g. Arithmetic Logic Units) could be used in the final design solutions.

Bambu: A modular framework for the high level synthesis of memory-intensive applications

This paper presents bambu, a modular framework for research on high-level synthesis currently under development at Politecnico di Milano. It can accept most of C constructs without requiring any three-state for their implementations by exploiting a novel and efficient memory architecture. It also allows the integration of floating-point units and thus it can deal with a wide range of data types. Finally, it allows to easily customize the synthesis flow (e.g., transformation passes, constraints, options, synthesis scripts) through an XML file and it automatically generates test-benches and validates the results against the corresponding software execution, supporting both ASIC and FPGA technologies.

HW/SW methodologies for synchronization in FPGA multiprocessors

odern Field Programmable Gate Arrays (FPGA) can be programmed with multiple soft-core processors. These solutions can be used for MultiProcessor Systems-on-Chip (MPSoCs) prototyping or even for final implementation. Nevertheless, efficient synchronization is required to guarantee performance in multiprocessing environments with the simple cores that do not support atomic instructions and are normally used in the standard FPGA toolchains. In this paper, we introduce two hardware synchronization modules for Xilinx MicroBlaze systems, with local polling or queuing mechanisms for locks and barriers, and present a comparison of these solutions to alternative designs.

A design methodology to implement memory accesses in high-level synthesis

Nowadays, the memory synthesis is becoming the main bottleneck for the generation of efficient hardware accelerators. This paper presents a design methodology to efficiently and automatically implement memory accesses in High-Level Synthesis. In particular, the approach starts from a behavioral specification (in pure C language) and a set of design constraints, such as the memory addresses where some of the data are stored. The methodology classifies which variables can be internally or externally allocated to the different modules to generate the proper architecture, fully supporting a wide range of C constructs, such as pointer arithmetic, function calls and array accesses. Moreover it allows to parallelize the accesses when the memory address is known at compile time, resulting in an efficient execution of the specification.

A runtime adaptive controller for supporting hardware components with variable latency

Nowadays, the design of hardware cores has to necessarily deal with unpredictable components, due to process variation or to the interaction with external modules (e.g., memories, sensors, IP cores). Adaptive systems are, thus, one of the most important solutions to substitute traditional approaches, based on analysis at design time, especially in critical environments. In this paper, we present an innovative lightweight controller architecture able to automatically adjust its behavior at run-time. It interacts with the surrounding environment by means of a simple token-based communication schema. We examine the capabilities of the proposed architectural model to adapt its behavior during the execution, compared to classical ones, such as the finite state machine.