Optimizing Data Pipelines for Machine Learning in Feature Stores

Author:

Liu Rui1,Park Kwanghyun2,Psallidas Fotis3,Zhu Xiaoyong3,Mo Jinghui4,Sen Rathijit3,Interlandi Matteo3,Karanasos Konstantinos5,Tian Yuanyuan3,Camacho-Rodríguez Jesús3

Affiliation:

1. University of Chicago

2. Yonsei University

3. Microsoft

4. LinkedIn

5. Meta

Abstract

Data pipelines (i.e., converting raw data to features) are critical for machine learning (ML) models, yet their development and management is time-consuming. Feature stores have recently emerged as a new "DBMS-for-ML" with the premise of enabling data scientists and engineers to define and manage their data pipelines. While current feature stores fulfill their promise from a functionality perspective, they are resource-hungry---with ample opportunities for implementing database-style optimizations to enhance their performance. In this paper, we propose a novel set of optimizations specifically targeted for point-in-time join, which is a critical operation in data pipelines. We implement these optimizations on top of Feathr: a widely-used feature store, and evaluate them on use cases from both the TPCx-AI benchmark and real-world online retail scenarios. Our thorough experimental analysis shows that our optimizations can accelerate data pipelines by up to 3× over state-of-the-art baselines.

Publisher

Association for Computing Machinery (ACM)

Subject

General Earth and Planetary Sciences,Water Science and Technology,Geography, Planning and Development

Reference61 articles.

1. 2019. Delta Lake. https://delta.io/. Accessed: 2023-02-23. 2019. Delta Lake. https://delta.io/. Accessed: 2023-02-23.

2. 2022. Amazon Redshift - Automated materialized views. https://docs.aws.amazon.com/redshift/latest/dg/materialized-view-auto-mv.html. Accessed: 2022-10-02. 2022. Amazon Redshift - Automated materialized views. https://docs.aws.amazon.com/redshift/latest/dg/materialized-view-auto-mv.html. Accessed: 2022-10-02.

3. 2022. Apache Spark. https://spark.apache.org/. Accessed: 2022-10-02. 2022. Apache Spark. https://spark.apache.org/. Accessed: 2022-10-02.

4. 2022. Apache Spark in Azure Synapse Analytics. https://learn.microsoft.com/azure/synapse-analytics/spark/apache-spark-overview. Accessed: 2022-10-02. 2022. Apache Spark in Azure Synapse Analytics. https://learn.microsoft.com/azure/synapse-analytics/spark/apache-spark-overview. Accessed: 2022-10-02.

5. 2022. Azure Blob Storage. https://azure.microsoft.com/en-us/products/storage/blobs. Accessed: 2022-10-02. 2022. Azure Blob Storage. https://azure.microsoft.com/en-us/products/storage/blobs. Accessed: 2022-10-02.

Cited by 1 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. The Hopsworks Feature Store for Machine Learning;Companion of the 2024 International Conference on Management of Data;2024-06-09

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3