Rui Su
Open-source advocate and co-founder of JuiceFS, a cloud-native distributed file system
Actions
Rui Su is the co-founder of Juicedata and creator of JuiceFS, an open-source distributed file system built on object storage. He has spent the past seven years building storage infrastructure for AI, analytics, and cloud-native environments, and leading JuiceFS from a proprietary product into a globally adopted Apache 2.0 project.
Rui speaks regularly on topics including AI/ML storage, Kubernetes-native data infrastructure, POSIX access over object storage, and open-source project strategy. He has presented at KubeCon 2023 Shanghai, SNIA SDC 2025, Community Over Code 2025, and KubeCon 2025 Atlanta.
Selected talks:
SNIA SDC 2025 — Why s3fs Fails in AI/ML and How to Achieve Scalable POSIX Access Anyway
https://www.youtube.com/watch?v=deQKBSAvt_M
Community Over Code 2025 — Why We Chose Openness: The Story Behind JuiceFS Switching to Apache 2.0
https://www.youtube.com/watch?v=DtyJHVz4CJE
KubeCon 2023 Shanghai — How to Accelerate Model Training and Eliminate the I/O Bottleneck for the Cloud (in Chinese)
https://www.youtube.com/watch?v=pK9pUxRpBSM
Scaling Autonomous-Driving R&D: A 100 PB, 100 Billion-File Open-Source Elastic File System on K8s
Autonomous-driving R&D pipelines generate petabytes of multi-modal sensor, simulation, and annotation data each month. Teams therefore need a storage layer that scales with their fleets—without forcing them to rewrite trusted tools.
In this talk we present a cloud-agnostic, open-source elastic file system already running in production at multiple OEMs and robo-taxi startups:
- Single namespace beyond 100 PB / 100 billion files — backed by commodity object storage yet mounted as a local POSIX volume.
- Thousands of Kubernetes nodes (mixed GPU/CPU) share the same dataset for training, validation, and replay with zero duplication.
- Multi-protocol endpoints (POSIX, S3, HDFS, WebDAV) let CV, mapping, and simulation teams keep their existing workflows.
- Hybrid & multi-cloud deployment—public cloud or on-prem with Ceph, enabling fully air-gapped SDV Labs.
We will walk through the high-level design patterns—stateless clients, scale-out metadata, transparent tiering—and share measured outcomes: saved millions in storage cost, scaled single-volume capacity 10×, with painless vendor portability. Attendees will leave with an opinionated reference diagram they can implement immediately.
Why We Replaced S3FS: Lessons from Building a Better Filesystem for AI Workloads on Kubernetes
Mounting S3-compatible storage via S3FS seems like an easy way to enable POSIX-like access in Kubernetes. But in real AI/ML workloads—e.g., training with PyTorch or TensorFlow—we hit major issues: crashes from incomplete writes, vanished checkpoints, inconsistent metadata, and unpredictable I/O latency.
This session shares our journey from debugging S3FS failures to deploying a scalable, POSIX-compliant file system that still leverages object storage. We’ll cover:
- Benchmarks comparing S3FS and a user-space distributed FS
- I/O traces showing metadata and small file pain points
- Key design decisions for compatibility and performance
- Kubernetes CSI and Operator integration for scale
- Lessons from running it on 1,000+ node AI training clusters
Ideal for platform engineers, MLOps, and Kubernetes architects seeking reliable, scalable storage for data-heavy workloads.
Why We Chose Openness: The Story Behind JuiceFS Switching to Apache 2.0
In 2022, to mark its first anniversary as an open source project, JuiceFS made a bold move: switching its license from AGPLv3 to Apache 2.0. At a time when many projects were going the opposite direction—adopting more restrictive licenses—we doubled down on openness. This lightning talk shares the thinking behind that decision, the risks we weighed, and the unexpected benefits it brought. I’ll also reflect on how this choice has shaped the growth and community of JuiceFS over the past few years, and what it means for sustainable open source in a commercial world.
The Three Things That Make or Break a Software Startup｜决定软件初创公司成败的三个关键因素
What really determines success or failure in a software startup? In this keynote, I’ll share the three pillars that matter most: a viable business model, clear product positioning, and sustainable user acquisition. These principles aren’t theoretical — they’re distilled from seven years of building JuiceFS, a distributed file system that went from closed-source to open-source, from SaaS-only to on-premises deployment, and from zero to adoption in some of the most demanding data infrastructures.
I’ll share hard-won lessons, strategic pivots, and industry shifts I’ve seen along the way. If you're building, investing in, or contributing to software ventures — especially open source — this talk will offer practical insights to navigate what truly matters.
是什么真正决定了一家软件初创公司的成败？在本次主题演讲中，我将分享最为重要的三大支柱：可行的商业模式、清晰的产品定位以及可持续的用户获取方式。这些原则并非空谈理论 —— 它们是从七年打造 JuiceFS 的实践中提炼而来的。JuiceFS 作为一款分布式文件系统，经历了从闭源到开源、从仅提供 SaaS 服务到支持本地部署的转变，最终从零开始，被一些要求极为严苛的数据基础设施所采用。
我会分享一路走来积累的宝贵经验、战略性转型以及观察到的行业变迁。如果你正在打造、投资软件企业，或是为其贡献力量 —— 尤其是开源领域的相关从业者，这场演讲将为你提供切实可行的见解，助你把握真正关键的要素。
KubeCon + CloudNativeCon North America 2025 Sessionize Event
Community Over Code Asia 2025 Sessionize Event
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top