论文标题
shisha：在线计划在异质体系结构上的CNN管道
Shisha: Online scheduling of CNN pipelines on heterogeneous architectures
论文作者
论文摘要
芯片已成为现代芯片设计中的一种常见方法。 chiplet在核心，内存子系统和互连水平上提高了产量和启用异质性。卷积神经网络（CNN）具有较高的计算，带宽和记忆能力要求，这是由于越来越大的权重。因此，为了利用基于chiplet的体系结构，必须根据计算资源之间的调度和工作负载分配来优化CNN。我们提出了Shisha，这是一种在线方法，用于生成和安排CNN Pullecal Pipeline on Chiplet Architectures。 Shisha针对计算性能和内存带宽方面的异质性，并通过快速的在线探索技术调整管道时间表。我们将Shisha与模拟退火，爬山和管道进行比较。与其他探索算法相比，平均而言，Shisha的收敛时间在Shisha中提高了约35倍。尽管进行了快速探索，但Shisha的解决方案通常比其他启发式探索算法要好。
Chiplets have become a common methodology in modern chip design. Chiplets improve yield and enable heterogeneity at the level of cores, memory subsystem and the interconnect. Convolutional Neural Networks (CNNs) have high computational, bandwidth and memory capacity requirements owing to the increasingly large amount of weights. Thus to exploit chiplet-based architectures, CNNs must be optimized in terms of scheduling and workload distribution among computing resources. We propose Shisha, an online approach to generate and schedule parallel CNN pipelines on chiplet architectures. Shisha targets heterogeneity in compute performance and memory bandwidth and tunes the pipeline schedule through a fast online exploration technique. We compare Shisha with Simulated Annealing, Hill Climbing and Pipe-Search. On average, the convergence time is improved by ~35x in Shisha compared to other exploration algorithms. Despite the quick exploration, Shisha's solution is often better than that of other heuristic exploration algorithms.
