一句话：分析 Google 消费级工作负载的数据移动开销，并证明存内计算卸载能显著降低能耗与执行时间。
消费设备（智能手机、平板、Chromebook、可穿戴设备）受电池容量与热功耗预算限制，能效至关重要。论文发现数据在内存系统与计算单元间移动的能耗和性能开销远高于计算本身。作者系统分析了四类广泛使用的 Google 消费级工作负载：Chrome 浏览器、TensorFlow Mobile、视频播放和视频捕获。结果表明，这些工作负载包含可贡献大量数据移动的简单原语与函数；在消费设备的面积和功耗约束下，将这些原语卸载到由简单核心或专用加速器组成的存内计算（PIM）逻辑是可行的。PIM 可消除大量数据移动，平均降低系统能耗 55.4%、执行时间 54.2%。
We are experiencing an explosive growth in the number of consumer devices, including smartphones, tablets, web-based computers such as Chromebooks, and wearable devices. For this class of devices, energy efficiency is a first-class concern due to the limited battery capacity and thermal power budget. We find that data movement is a major contributor to the total system energy and execution time in consumer devices. The energy and performance costs of moving data between the memory system and the compute units are significantly higher than the costs of computation. As a result, addressing data movement is crucial for consumer devices. In this work, we comprehensively analyze the energy and performance impact of data movement for several widely-used Google consumer workloads: (1) the Chrome web browser; (2) TensorFlow Mobile, Google's machine learning framework; (3) video playback, and (4) video capture, both of which are used in many video services such as YouTube and Google Hangouts. We find that processing-in-memory (PIM) can significantly reduce data movement for all of these workloads, by performing part of the computation close to memory. Each workload contains simple primitives and functions that contribute to a significant amount of the overall data movement. We investigate whether these primitives and functions are feasible to implement using PIM, given the limited area and power constraints of consumer devices. Our analysis shows that offloading these primitives to PIM logic, consisting of either simple cores or specialized accelerators, eliminates a large amount of data movement, and significantly reduces total system energy (by an average of 55.4% across the workloads) and execution time (by an average of 54.2%).
DOI 原文 ·
@inproceedings{paperbot4292,
title = {Google Workloads for Consumer Devices},
author = {Amirali Boroumand and Saugata Ghose and Youngsok Kim and Rachata Ausavarungnirun and Eric Shiu and Rahul Thakur and Daehyun Kim and Aki Kuusela and Allan Knies and Parthasarathy Ranganathan and Onur Mutlu},
booktitle = {ASPLOS},
year = {2018},
doi = {10.1145/3173162.3173177}
}