Exploring Next‑Generation Big Data Storage Engines: Paimon, JuiceFS, Databend, and CloudFS

The big‑data ecosystem has evolved over more than a decade, moving from on‑premise data centers to cloud‑based solutions. Early storage systems like HDFS gave way to object‑storage‑centric, compute‑storage separation architectures, and newer concepts such as Data Lake and Lakehouse. In this summit, several speakers present their explorations of next‑generation storage engines including Paimon, JuiceFS, Databend, and CloudFS.

Event Details: The summit took place online from 09:00 to 12:30 on August 5‑6, 2023, with live streaming and a QR‑code for registration.

Speaker: Su Rui (Juicedata Partner) Su Rui has been involved in JuiceFS product, market, and open‑source community building since 2017, with 16 years of experience across software, internet, and NGO sectors.

Speaker: Li Ming (Senior R&D Engineer, Duodian DMALL Data Platform) Li Ming is responsible for the company's cloud‑native big‑data architecture and data‑base new features, focusing on unified SQL gateways, distributed file storage, high‑performance computing, and data security. He contributes to open‑source projects such as Apache Kyuubi, JuiceFS, Apache Celeborn, and Trino.

Talk: "Duodian DMALL × JuiceFS: Storage Architecture Exploration under Big‑Data Compute‑Storage Separation" The presentation covers challenges of integrated compute‑storage, the evolution to a separated architecture, and deep‑dive practices of JuiceFS.

Speaker: Guo Jun (Head of Big‑Data File Storage, Volcano Engine) With experience at Microsoft Azure Stack, Huawei Storage, and Alibaba Cloud, Guo Jun now leads Volcano Engine’s CloudFS, delivering simple, stable, and multi‑scenario compatible storage acceleration services.

Talk: "Volcano Engine Cloud‑Native Storage Acceleration Practice" The talk explains the cloud‑native K8s environment for machine learning and data lake workloads, the design of CloudFS, and multi‑scenario acceleration solutions, followed by audience benefits.

Speaker: Wang Gang (Senior Data Engineer, Autohome) Wang has rebuilt logging platforms, built real‑time computing pipelines with Apache Flink, and pioneered lake‑warehouse integration using Apache Iceberg and Milvus. He now explores production use of Apache Paimon.

Talk: "Autohome Production Practice with Paimon" The agenda includes an overview of Paimon, its core functions and architecture, and detailed case studies of its adoption at Autohome.

Speaker: Wu Bingxi (Co‑Founder & Architect, Databend Labs) Wu focuses on cloud‑native big‑data analytics and is familiar with the MySQL ecosystem.

Talk: "Serverless, Cross‑IDC Data Analytics Platform on the Cloud" The session discusses current big‑data characteristics and cloud challenges, optimal cloud resource strategies, design of a serverless, storage‑separated data warehouse, and practical experiences with Databend Cloud, ending with a Q&A.

All sessions emphasize practical insights, architectural design, and audience takeaways such as understanding cloud‑native storage acceleration, Paimon’s production usage, and serverless big‑data cost‑efficiency.