一句话：指出大 NN 扩展加速器会加剧片上 SRAM 与片外 DRAM 的存储成本和能耗问题，并以 TETRIS 为题研究该瓶颈。
本文关注深度神经网络（NN）加速器在规模扩展时面临的存储系统开销。虽然 NN 的高精度推动了性能提升两个数量级的加速器发展，但用越来越大的 NN 继续扩展这些加速器，会加剧其存储系统的成本与能耗负担，包括片上 SRAM 缓冲和片外 DRAM 通道。论文以 TETRIS 为题，针对这一瓶颈展开研究；然而给出的摘要仅陈述了问题背景，未说明具体方法、系统设计或实验结果，因此无法从现有信息判断其技术路线与量化收益。
The high accuracy of deep neural networks (NNs) has led to the development of NN accelerators that improve performance by two orders of magnitude. However, scaling these accelerators for higher performance with increasingly larger NNs exacerbates the cost and energy overheads of their memory systems, including the on-chip SRAM buffers and the off-chip DRAM channels.