一句话：Homunculus 自动为数据中心网络生成满足约束的高效数据平面机器学习流水线。
过去十年，网络中对机器学习（ML）应用的支持显著提升；公开数据集与可编程交换结构（包括用于编程它们的底层语言）为在网络内 ML 部署提供了全栈能力。然而，相关工具种类繁多，且需在满足吞吐、延迟等网络约束的同时完成 ML 模型设计与超参数调优等复杂优化任务，这迫使网络运营者同时成为 ML、网络设计与可编程硬件专家。Homunculus 针对这一痛点，面向数据中心网络自动生成高效的数据平面 ML 流水线，以降低手工设计、调优与部署负担。摘要未给出具体方法与实验评估信息。
Support for Machine Learning (ML) applications in networking has significantly improved over the last decade. The availability of public datasets and programmable switching fabrics (including low-level languages to program them) presents a full-stack to the programmer for deploying in-network ML. However, the diversity of tools involved, coupled with complex optimization tasks of ML model design and hyperparameter tuning while complying with the network constraints (like throughput and latency), puts the onus on the network operator to be an expert in ML, network design, and programmable hardware.
@inproceedings{paperbot10055,
title = {Homunculus: Auto-Generating Efficient Data-Plane ML Pipelines for Datacenter Networks},
author = {Tushar Swamy and Annus Zulfiqar and Luigi Nardi and Muhammad Shahbaz and Kunle Olukotun},
booktitle = {ASPLOS},
year = {2023},
doi = {10.1145/3582016.3582022}
}