Search

MPSoCs with hierarchical communication infrastructures are promising architectures for low power embedded systems. Multiple CPU clusters are coupled using an Network-on-Chip (NoC). Our CoreVA-MPSoC targets streaming applications in embedded systems, like signal and video processing. In this work we introduce a tightly coupled shared data memory to each CPU cluster, which can be accessed by all CPUs of a cluster and the NoC with low latency. The main focus is the comparison of different memory architectures and their connection to the NoC. We analyze memory architectures with local data memory only, shared data memory only, and a hybrid architecture integrating both. Implementation results are presented for a 28 nm FD-SOI standard cell technology. A CPU cluster with shared memory shows similar area requirements compared to the local memory architecture. We use post place and route simulations for precise analysis of energy consumption on both cluster and NoC level using the different memory architectures. An architecture with shared data memory shows best performance results in combination with a high resource efficiency. On average, the use of shared memory shows a 17.2 percent higher throughput for a benchmark suite of 10 applications compared to the use of local memory only.

Embedded streaming applications are facing increasingly demanding performance requirements in terms of throughput. A common mechanism for providing high compute power with a low energy budget is to use a very large number of low-power cores, often in the form of a Massively Parallel System on Chip (MPSoC). The challenge with programming such massively parallel systems is deciding how to optimally map the computation to individual cores for maximizing throughput. In this work we present an automatic parallelizing compiler for the StreamIt programming language that efficiently and effectively maps computation to individual cores. The compiler must be both effective, meaning that it does a good job of optimizing for throughput; but also efficient, in that the time taken to find such a mapping must scale well as the number of cores and size of the Stream program increases. We improve on previous work that used Integer Linear Programming (ILP) to map StreamIT programs to multicore systems by formulating the mapping problem in a different way using mostly real rather than integer variables. Using so called Mixed Integer Linear Programming (MILP) dramatically reduces the cost compared to standard ILP. This alternative formulation creates what we call an optimistic solution that we then need to adjust slightly to obtain a final feasible solution. We show that this new approach is always close, if not better in terms of effectiveness, while being dramatically better in terms of scalability and efficiency.

In large-scaled Multiprocessor System-on-Chips (MPSoCs) the Globally Asynchronous Locally Synchronous (GALS) design paradigm is widely-used to minimize power consumption. Especially Network-on-Chips (NoCs) utilize several GALS approaches to connect hundreds of CPU cores on a single chip. However, improvements in modern design tools, like the clock concurrent optimization flow (CCOpt-flow), require to rethink the benefit of these GALS-based approach. Therefore, the main focus of this work is the comparison of a synchronous (using CCOpt-flow), a multi-synchronous (mesochronous) and a fully asynchronous NoC design for the usage in hierarchical MPSoCs. In hierarchical MPSoCs multiple CPU cores are tightly coupled to clusters, which in turn are connected via a NoC. As a result, wire length for establishing NoC links between the clusters increases, making a NoC design more challenging. Implementation results for a 28nm FD-SOI standard cell technology have shown that the CCOpt-flow still allows to scale a fully synchronous MPSoCs, with slight drawbacks in maximum clock frequency. Nevertheless, the asynchronous NoC outperforms the other two approaches with respect to performance, power and area requirements. The asynchronous NoC leads to a 3:1% smaller MPSoC design. In addition, the proposed asynchronous NoC router consumes only 53% of the power of a mesochronous router during active communication.

Parallel programming and effective partitioning of applications for embedded many-core architectures requires optimization algorithms. However, these algorithms have to quickly evaluate thousands of different partitions. We present a fast performance estimator embedded in a parallelizing compiler for streaming applications. The estimator combines a single execution-based simulation and an analytic approach. Experimental results demonstrate that the estimator has a mean error of 2.6% and computes its estimation 2848 times faster compared to a cycle accurate simulator.

Network Interfaces (NIs) are used in Multiprocessor System-on-Chips (MPSoCs) to connect CPUs to a packet switched Network-on-Chip. In this work we introduce a new NI architecture for our hierarchical CoreVA-MPSoC. The CoreVA-MPSoC targets streaming applications in embedded systems. The main contribution of this paper is a system-level analysis of different NI configurations, considering both software and hardware costs for NoC communication. Different configurations of the NI are compared using a benchmark suite of 10 streaming applications. The best performing NI configuration shows an average speedup of 20 for a CoreVA-MPSoC with 32 CPUs compared to a single CPU. Furthermore, we present physical implementation results using a 28 nm FD-SOI standard cell technology. A hierarchical MPSoC with 8 CPU clusters and 4 CPUs in each cluster running at 800 MHz requires an area of 4.56 mm².

Tightly coupling CPUs within clusters allows for low latency, high bandwidth communication in MPSoCs. Our CoreVA-MPSoC integrates multiple clusters using an on-chip network. In this work we introduce a shared L1 data memory that can be accessed by all CPUs of a cluster with low latency. A Mesh-of-Trees (MoT) and a crossbar as topology for the shared memory interconnect are presented. A cluster that integrates the shared L1 memory is compared with an architecture that features an AXI interconnect and a local L1 data memory for each CPU. In addition, we consider an architecture that integrates both. We present implementation results using a 28nm FD-SOI standard cell technology. The shared L1 memory shows similar area results compared to the local memory architecture. Place and route results of a cluster with 8 CPUs, 128kB local-and 128kB shared L1 data memory divided into 16 memory banks show a frequency of 728MHz and an area of 1.77mm2. To map programs to the different CPU cluster configurations a compiler for streaming applications is used. An architecture with both local and shared L1 data memory and 4 memory banks shows best performance results in combination with a high resource efficiency.

Energy efficient embedded computing enables new application scenarios in mobile devices like software-defined radio and video processing. The hierarchical multiprocessor considered in this work may contain dozens or hundreds of resource efficient VLIW CPUs. Programming this number of CPU cores is a complex task requiring compiler support. The stream programming paradigm provides beneficial properties that help to support automatic partitioning. This work describes a compiler for streaming applications targeting the self-build hierarchical CoreVA-MPSoC multiprocessor platform. The compiler is supported by a programming model that is tailored to fit the streaming programming paradigm. We present a novel simulated-annealing (SA) based partitioning algorithm, called Smart SA. The overall speedup of Smart SA is 12.84 for an MPSoC with 16 CPU cores compared to a single CPU implementation. Comparison with a state of the art partitioning algorithm shows an average performance improvement of 34.07%.