Hierarchical Dataflow Model for efficient programming of clustered manycore processors

Multiprocessor System-on-Chip (MPSoC) designs offer a lot of computational power assembled in a compact design. The computing power of MPSoCs can be further augmented by adding heterogeneous processing elements, e.g. massively parallel processor arrays (MPPA) and specialized hardware with instruction-set extensions. However, the presence of multiple processing elements (PEs) with different characteristics raises issues related to programming and application mapping. The conventional approach used for programming heterogeneous MPSoCs results in a static mapping of various parts of the application to different PE types, based on the nature of the algorithm and the structure of the PEs. Yet, such a mapping scheme independent of the instantaneous load on the PEs may lead to under-utilization of some type of PEs while overloading others. We investigate the benefits of a resource-aware programming model called Invasive Computing for dynamically mapping image processing applications to different types of PEs available on a heterogeneous MPSoC. A case study of visual object recognition is presented, including Harris corner detection and SIFT feature matching. Results indicate that resource-aware programming helps to predict the latency of the application program along with better overall workload distribution within the heterogeneous MPSoC.

Embedded System plays a vital role in consumer Industry. Complex applications need systems which contains multiple heterogeneous processors, running in parallel to speed up the system. Also due to area constraints, the processors are evolved in a single System on Chip called Multiprocessor System on Chip (MPSoC). The system should be reusable and debuggable, hence the designers designed and developed Reconfigurable MPSoCs rather than Application Specific Integrated Circuits (ASIC) in Field Programmable Gate Arrays (FPGA). Multiprocessor System on Chip (MPSoC) platform plays a vital role in parallel processor architecture design. However the growth of number of processing elements in one chip, task decomposition and scheduling become major bottlenecks of MPSoC architecture. To execute the applications, the application software is splitted as tasks and mapped to the different available processors and scheduled the tasks as when to execute in the available processors when the resources are ready. Selection of most suitable candidates for execution in a particular processor is very much important. Hardware related tasks are executed in different hardware accelerators and software tasks in processors. The area occupied by the schedulers in memory is more in internal memory. For scheduling these tasks, a programmable hardware is developed as hardware scheduler in the reconfigurable MPSoC using NIOS II processor. The algorithm for optimized scheduling in the target architecture is proposed. The literature survey is made with the hardware scheduler and new target MPSoC architecture. Quartus II version 12.1 and SOPC Builder are used to configure the NIOS II processer. Nios II EDS software tool has been used to build the application code.

The design of a Multiprocessor System-on-Chip (MPSoC) is a complex task, including steps as application development, platform configuration, code generation, task mapping onto the platform and debugging. An integrated environment covering most of these steps is a gap in the literature. The present work first details an MPSoC architecture, which supports the execution of distributed applications, including an operating system enabling multitask execution at each processing element. The MPSoC is heterogeneous, due to the support to different processor architectures. Then, a framework able to cover the design steps previously mentioned is presented. The framework enables the design space exploration for applications to be executed in the MPSoC, varying for example the number and type of processors, the memory size, the task mapping. Results demonstrate the correct operation for different MPSoC configurations, generated from the proposed framework. Such open-source framework enables the research community to investigate new subjects related to MPSoC and Network on Chip (NoC) design, as well as evaluate distributed applications in a multiprocessor environment.

Multiprocessor Systems-on-Chip (MPSoCs) are a promising solution to fulfill the requirements of high performance computing applications, such as image processing or bioinformatics. Many different MPSoC architectures are introduced by industry and academics, such as Intel's Single Cloud Computer (SCC) or IBM's Cell. To achieve a high efficiency on such MPSoCs a high computing performance in the processing elements (PE) is not the only parameter. The bandwidth of the memory and especially the on-chip communication infrastructure are extremely important to achieve the requirements of the applications. Furthermore, for embedded System-on-Chip solutions, flexibility is extremely important for providing a good tradeoff between performance and power consumption. Especially, for general purpose MPSoCs, designed for a wide range of applications with different communication demands, this flexibility is essential for a good energy efficiency. A simple bus or Network-on-Chip (NoC) with fixed bandwidth and communication paradigm is not sufficient in this varying application scenario and its design space. In this paper novel features for the heterogeneous and runtime adaptive Star-Wheels Network-on-Chip are presented. This Network-on-Chip is used within a runtime adaptive MPSoC, which enables to tailor processors and accelerators to the requirements of the applications. For optimizing the Star-Wheels Network-on-Chip and its communication protocol at runtime to the application requirement the following novel features have been added: the combination of the circuit- and the packet-switched communication paradigm and an algorithm-based placement of the processing elements within the NoC. The integration of the novel algorithm into the design methodology and the exploitation of the novel features within the special purpose runtime operating system of a runtime adaptive MPSoC are also introduced.

Nowadays, aggressive scaling in integrated circuits brings out new challenges such as increase in power density, temperature, and process variation in designing Multiprocessor Systems-on-Chip (MPSoC) employed in embedded systems. While most of the previous works attempt to mitigate the process variation effects in system design level, the eventual design still is inefficient and suffers from the variability of frequency and leakage power of processors in a MPSoC. In this paper, we formulate a MILP problem for variation-aware task assignment and scheduling to optimize power consumption while meeting the real-time constraints. To capture stochastic behavior of process variation, we employ chance-constrained programming technique to turn the problem into a corresponding stochastic optimization one that can be solved by typical solvers. Extensive experiments using E3S benchmarks have been carried out and the obtained results of the proposed method evince improvements compared to the baseline method in terms of performance-yield and run-time.

Multiprocessor System-on-Chip (MPSoC) based on Network-on-Chip (NoC) integrates a large amount of Processor Elements (PEs) to fulfill the performance requirements of several applications. These applications are composed of a set of intercommunicating tasks, which are dynamically mapped onto PEs of the target architecture. However, the efficient task-mapping requires some previous steps, among them partitioning, which organizes tasks considering their interaction before applying a mapping process. This paper introduces Partition Reduce (PR) — a task partitioning approach based on the MapReduce algorithm targeting homogeneous NoC based MPSoCs. We analyze the efficiency of PR for energy consumption (EC) minimization and load balance (LB). The results obtained from a set of experiments, with large number of tasks, demonstrate that PR is more effective on processing time and result quality when compared to the classic Simulated Annealing (SA). In addition, PR produces partitions with low energy consumption and rigorous load balance.

With technology scaling, Multiprocessor System on Chip (MPSoC) which consist of multiple processors connected via a Network on Chip (NoC) have become prevalent. Applications are mapped to MPSoC's by representing it in the form of a task graph. Task scheduling involves mapping task to processor cores so as to meet the deadline. For a given deadline, slack at each node is defined by the amount of time by which a task execution can be delayed without missing the deadline. With increase in the number of cores and high application parallelism, NoC is becoming a bottleneck due to the presence of large number of concurrent communications. Increasing network resources (links and routers) reduces the communication time but the area and power goes up. In this paper we present an application aware heuristic to synthesize a minimal network connecting a set of cores in an MPSoC in the presence of hard deadlines. Our approach is based on modeling communication between a pair of processors as tasks known as “Comtasks”. The network is generated by “scheduling” these comtasks onto a set of routers so as to obtain a network with minimum area which fills up the available slack. Moreover, we also identify the set of overlapping comtasks and generate a minimal network to allow the maximum set of overlapping comtasks to execute concurrently. We compared our approach with a greedy network generation heuristic and the results show 80% benefit in the router area.

Nowadays, the growing complexity in a wide variety of applications has led to the use of Multiprocessor Systems-on-Chip (MPSoCs). Networks-on-Chip (NoCs) have emerged as a scalable intra-chip communication technology for a high number of processing elements. However, the evolution of MPSoCs shows increasing communication requirements due to the growing number of PEs. This paper presents a novel Network-on-Chip microarchitecture that combines data transfers with data stream processing to exploit efficiently the transmission time. Data stream processing is enabled by a lightweight processing unit that is located within the transmission channels of the NoC. Thus, data can be processed by operations either in a processing element such as a MicroBlaze processor or inside the NoC while it is transferred. Operations executed by the NoC reduce the workload of the processing elements. In addition, the transmission time will be efficiently used. The novel microarchitecture is evaluated and compared to a state-of-the-art router on a Xilinx Zynq FPGA.

The progressive maturity of VLSI manufacturing technology is helping in integrating more and more processing elements and memory units on a single die to form a Multiprocessor System-On-Chip (MPSoC). Network-on-Chip (NoC) is adopted as communication backbone for most of these modern day multiprocessor systems. As complexity of these system scales, there has been a growing concern on the dependability of these processing and communication elements. In this paper, we propose a centralized hardware fault-tolerant network interface (NI) for NoCs based on spatial division multiplexing. Experiments show that the proposed design has better throughput than a non fault-tolerant design with only 18% area overhead. We also introduce an area optimized distributed fault-tolerant NI architecture which provides 50% more throughput than the centralized design for high fault rates.

The heterogeneous multicore system integrated with FPGA is a kind of Multi-Processor Systems-on-Chip(MPSoC) that can achieve both efficiency and flexibility. FPGA uses hardware resource instead of instruction sets to process tasks. The resource on FPGA is limited which should be considered when scheduling tasks. We propose a static task scheduling algorithm targeting this kind of system. The aim is to minimize the executing time of an application as well as considering FPGA resource limit. In our genetic algorithm-based method, a chromosome consists of computing units where tasks are assigned. When generating the initial population, some tasks are assigned to FPGA, considering FPGA resource limit. We have modified the crossover operator and mutation operator to ensure that the FPGA resource used implied in chromosomes will not exceed the FPGA resource limit. Task scheduling is completed in the chromosome evaluation stage. Through the genetic algorithm, the improved task assignment and schedule sequence are obtained. The experiments on random graph applications and two real-world applications show that our method has achieved better performance than existing works.

All computing platforms, from mobile to supercomputers, are becoming more and more heterogeneous and massively parallel. While they can provide higher power efficiency and computation throughput, effective and confident use of these systems always requires knowledge about low-level programming. The average time necessary to develop and optimize a design on heterogeneous platforms is higher and higher compared to typical homogeneous systems. Dataflow models of computation (MoC) are quickly becoming the common practice in heterogeneous systems development. In domains such as signal processing and multimedia communication, dataflow MoCs have become accepted as standard. However, the shift from a sequential and architecture-specific MoC to a dataflow MoC still uncovers several programming and development challenges. The Cal Actor Language (CAL) is a recently-specified dataflow and actor-based language capable of concisely expressing complex and general purpose parallel applications. However, design tools supporting this language are generally not adequate to fully exploit its features and expressiveness power. In fact, they generally restrict its MoC in order to reduce the design space exploration (DSE) effort. The objective of this thesis is to provide a DSE methodology where all the features of CAL and dynamic dataflow MoCs can be exploited in a more general and effective manner. This dissertation illustrates a novel profiling, analysis and performance estimation methodology for the DSE of dynamic dataflow programs. The main research contributions of this thesis are: the formalization of a graph-based representation of the program execution called an execution trace graph (ETG); the formalization of a systematic methodology for profiling generic dynamic dataflow programs through their code interpretation; the formalization of a complete DSE methodology for dynamic dataflow programs in order to efficiently identify close-to-optimal design points according to various and tailored performance merit functions. In particular, the following design space optimization problems for dynamic dataflow programs are addressed: the analysis of the hotspots and the algorithmic bottlenecks of a parallel program; the bounding and optimization of the buffer size configuration for complex designs; the dynamic power dissipation minimization of programs implemented in multi-clock domain architecture. Furthermore, theoretical concepts like the design space critical path and the potential speedup of a dataflow application have been defined and revisited, respectively. The thesis also presents a DSE framework developed in order to demonstrate the effectiveness of this design methodology.

Over the years, system on chip (SoC) designs have evolved extensively sophisticated in order to fulfil the need of increasing complexity of running applications driven by the advancement of VLSI technology. The next generation multiprocessor system on chip (MPSoC) integrates hundreds of processing elements on a single chip, expected to achieve high performance, low latency, and low-power consumption. Tiled chip multicore processors (TCMP) with network on chip (NoC) have become a foundation for the computation critical embedded and real-time systems. Faster time-to-market restraint and business competition compelled manufacturers to look for the prospects of manufacturing SoCs integrated with various third party intellectual property (3PIP). Usage of 3PIP gave rise to exploit the underlying interconnect while adding some unwanted malicious circuit known as hardware trojan (HT), making NoC vulnerable to get attacked. A tiniest manipulation of any communication attribute by HT can degrade the overall behaviour of the system significantly while impacting NoC performance metrics. In this paper, we present one of such study considering HT malign behaviour of manipulating the output port of each incoming flit from specific ports once genuine route computation takes place and re-directs all to one port causing disruption in tile communication.The proposed HT is intermittent in nature and activates for few cycles. We study the behaviour of proposed confluence attack and analyze its impact over network level. The empirical evaluation exhibits the misbehaviour of the packet communication in terms of wrong usage of VCs and extra hop distance overhead. To validate the proposed work, various performance metrics like buffer utilization, virtual channel utilization, number of flits processed, link utilization, and the like are analyzed. We also show that such HTs are difficult to detect due to marginal increase at malicious port traffic leaving no trace of malicious conduct.

Inter-core communication introduces overheads in task schedules on Multiprocessor System-on-Chips (MPSoCs). Inter-core communication overhead not only negatively impacts the timing performance but also significantly degrades the memory usage for streaming applications running on MPSoC architectures. By minimizing inter-core communication overhead, a shorter period can be applied and system performance (e.g., throughput, memory usage) can be improved. In this paper, we focus on solving the problem of minimizing inter-core communication overhead for streaming applications on bus-based MPSoCs. The objective is to minimize inter-core communication overhead while minimizing the overall memory usage. To solve the problem, we first let tasks with intra-period data dependencies transform to inter-period data dependencies so as to overlap the execution of computation and inter-core communication tasks. By doing this, inter-core communication overhead can be effectively removed. To minimize the overall memory usage, we then perform schedulability analysis and obtain the bounds of the times needed to reschedule each task. Based on the schedulability analysis, we formulate the scheduling problem as an integer linear programming (ILP) model and obtain an optimal schedule. In addition, we propose a heuristic approach to efficiently obtain a near-optimal solution. We conduct experiments on a set of benchmarks from both real-life streaming applications and synthetic task graphs. The experimental results show that the proposed approach can significantly reduce the schedule length and improve the memory usage compared with the previous work.

Multiprocessor system-on-chip (MPSoC) platforms have been emerging as the main solution to cope with processor frequency ceiling and power density issues while still improving performances. Then, network-on-chip (NoC) has been adopted to provide the increasing number of processors with the required communication bandwidth as well as with the necessary flexibility. Video processing and streaming applications are adopting dynamic dataflow model of computation as the need for high performance parallel computing is growing. Dataflow applications executed on modern MPSoC-based architectures are becoming increasingly dynamic and more data-dependent. Different tasks execute concurrently with significant modifications in their workloads and resource demanding over time depending on the input data. Hence, adopting any static or offline dynamic scheduling for mapping tasks will not cope with the computation variations. This article introduces an original run-time mapping algorithm based on the Move Based (MB) method targeting a dedicated heterogeneous NoC-based MPSoC architecture to achieve workload balancing and optimized communication traffic. The performance of the proposed algorithm is verified by conducting cycle-accurate SystemC simulations of the adopted NoC implementing a real MPEG4-SP decoder. The obtained results reveal the effectiveness of our proposed algorithm. For various real-life videos, the proposed algorithm systematically succeeded to enhance significantly the performance.

Dataflow formalisms have provided designers of digital signal processing systems with analysis and optimizations for many years. As system complexity increases, designers are relying on more types of dataflow models to describe applications while retaining these implementation benefits. The semantic range of DSP-oriented dataflow models has expanded to cover heterogeneous models and dynamic applications, but efficient design, simulation, and scheduling of such applications has not. To facilitate implementing heterogeneous applications, we utilize a new dataflow model of computation and show how actors designed in other dataflow models are directly supported by this framework, allowing system designers to immediately compose and simulate actors from different models. Using an example, we show how this approach can be applied to quickly describe and functionally simulate a heterogeneous dataflow-based application such that a designer may analyze and tune trade-offs among different models and schedules for simulation time, memory consumption, and schedule size.KeywordsDataflowHeterogeneousSignal Processing