No abstract

No abstract

Neuromorphic hardware, the new generation of non-von Neumann computing system, implements spiking neurons and synapses to spiking neural network (SNN)-based applications. The energy-efficient property makes the neuromorphic hardware suitable for power-constrained environments where sensors and edge nodes of the internet of things (IoT) work. The mapping of SNNs onto neuromorphic hardware is challenging because a non-optimized mapping may result in a high network-on-chip (NoC) latency and energy consumption. In this paper, we propose NeuMap, a simple and fast toolchain, to map SNNs onto the multicore neuromorphic hardware. NeuMap first obtains the communication patterns of an SNN by calculation that simplifies the mapping process. Then, NeuMap exploits localized connections, divides the adjacent layers into a sub-network, and partitions each sub-network into multiple clusters while meeting the hardware resource constraints. Finally, we employ a meta-heuristics algorithm to search for the best cluster-to-core mapping scheme in the reduced searching space. We conduct experiments using six realistic SNN-based applications to evaluate NeuMap and two prior works (SpiNeMap and SNEAP). The experimental results show that, compared to SpiNeMap and SNEAP, NeuMap reduces the average energy consumption by 84% and 17% and has 55% and 12% lower spike latency, respectively.

No abstract

Neuromorphic processors, the new generation of brain-inspired non-von Neumann computing systems, are developed to better support the execution of spiking neural networks (SNNs). The neuromorphic processor typically consists of multiple cores and adopts the Network-on-Chip (NoC) as the communication framework. However, an unoptimized mapping of SNNs onto the neuromorphic processor results in lots of spike messages on NoC, which increases the energy consumption and spike latency on NoC. Addressing this problem, we present a fast toolchain, NeuToMa, to map SNNs onto the neuromorphic processor. NeuToMa exploits the global topology of SNNs and uses the group optimization strategy to partition SNNs into multiple clusters, significantly reducing the NoC traffic. Then, NeuToMa dispatches the clusters to neuromorphic cores, minimizing the average hop of spike messages and balancing the NoC workload. The experimental results show that compared with the state-of-the-art technique, NeuToMa reduces the spike latency and energy consumption by up to 55% and 86%, respectively.