An Adaptive Memory Management Strategy Towards Energy Efficient Machine Inference in Event-Driven Neuromorphic Accelerators
An Adaptive Memory Management Strategy Towards Energy Efficient Machine Inference in Event-Driven Neuromorphic Accelerators
复制标题
DOI：
10.1109/asap.2019.000-2
复制
发表时间：
2019-07
期刊：
影响因子：
--
通讯作者：
Saunak Saha;Henry Duwe;Joseph Zambreno
中科院分区：
文献类型：
--
作者：
Saunak Saha;Henry Duwe;Joseph Zambreno
尖峰神经网络是在低功率嵌入和物联网设备中用于边缘处理的经典神经网络的可行替代方案。为了获得收益，倾向于支持深网的神经形态网络加速器仍然必须花费大量的精力来从较大的遥控内存中获取突触状态。由于这些网络中的本地计算是事件驱动的，因此内存成为系统能源消耗的主要部分。在本文中，我们探讨了各种数据重用机会，可以帮助减轻冗余流量以检索神经元元数据和突触后的重量。我们描述了Cynapse，这是一个基线神经处理单元及其随附的软件模拟，作为用于探索各个级别的一般模板。然后，我们研究了具有明显不同的拓扑和活动的三个尖峰神经网络基准的内存访问模式。通过详细的对记忆流量中的地方的研究，我们建立了阻碍传统的缓存管理理念在这些应用中有效工作的因素。为此，我们提出并评估特定领域的管理策略，该政策利用了基于队列的事件驱动的仿真框架中事件的前瞻性可见性。随后，我们提出了网络自适应增强功能，以使其对网络变化具有鲁棒性。结果，我们比传统的替代政策降低了系统功耗13-44％。
Spiking neural networks are viable alternatives to classical neural networks for edge processing in low-power embedded and IoT devices. To reap their benefits, neuromorphic network accelerators that tend to support deep networks still have to expend great effort in fetching synaptic states from a large remote memory. Since local computation in these networks is event-driven, memory becomes the major part of the system's energy consumption. In this paper, we explore various opportunities of data reuse that can help mitigate the redundant traffic for retrieval of neuron meta-data and post-synaptic weights. We describe CyNAPSE, a baseline neural processing unit and its accompanying software simulation as a general template for exploration on various levels. We then investigate the memory access patterns of three spiking neural network benchmarks that have significantly different topology and activity. With a detailed study of locality in memory traffic, we establish the factors that hinder conventional cache management philosophies from working efficiently for these applications. To that end, we propose and evaluate a domain-specific management policy that takes advantage of the forward visibility of events in a queue-based event-driven simulation framework. Subsequently, we propose network-adaptive enhancements to make it robust to network variations. As a result, we achieve 13-44% reduction in system power consumption and 8-23% improvement over conventional replacement policies.