As 2024 comes to a close, it marks the fourth year of JuiceFS Community Edition and the eighth year of its Enterprise Edition. This year has been a remarkable journey for us, with our community experiencing exponential growth. Our GitHub stars surpassed 11.1k, and all usage metrics increased by more than 100%. Notably, the total number of file systems grew by 8.5 times compared to the previous year.
With the fast-paced development of artificial intelligence (AI) technologies in 2024, AI is becoming the new infrastructure for enterprises, driving a shift towards AI-native architectures. This transformation has significantly boosted the JuiceFS community growth.
The proportion of users using JuiceFS for AI scenarios has surged, spanning industries such as education, manufacturing, wind power, and biotechnology. In the emerging AI sector, in addition to foundational model research companies, including China’s leading large language model (LLM) enterprises like MiniMax, ChatGLM, and StepFun, many small and medium-sized AI service software companies have joined the community, becoming a key user group.
Let’s take a closer look at JuiceFS’ achievements in 2024.
Product advancements: Enterprise-grade management, multi-cloud architecture, and large-scale AI scenarios
JuiceFS Community Edition
In 2024, JuiceFS Community Edition released 9 updates, maintaining its cadence of one major release per year. The 1.2 version was the third major release since its open-sourcing in 2021, delivering significant enhancements for enterprise-grade access management and usability in large-scale data scenarios. Key features include:
- Support for POSIX ACL
- Smooth upgrades
- Optimized S3 Gateway
- Enhanced
juicefs sync
JuiceFS CSI Driver
JuiceFS CSI Driver saw 16 releases in 2024, focusing on stability and usability in Kubernetes environments. Key updates include:
- Smooth upgrades
- Unified configuration
- The
kubectlplugin - Cache Group Operator
JuiceFS Enterprise Edition
JuiceFS Enterprise Edition was designed for high-performance computing with massive file systems. It underwent substantial optimization for multi-cloud architecture, AI ecosystem compatibility, and stability in large-scale AI scenarios. The v5.1 release introduced several key features:
- Write support for mirror clusters
- Python SDK
- Distributed cache replication
- S3 Gateway account management
In the next version, we’ll tackle even more extreme scenarios, focusing on:
- Distributed cache performance optimization
- Hotspot auto-balancing
- High-performance FUSE improvements
AI-powered community growth: Engagement and activity
In its fourth year as an open-source project, JuiceFS Community Edition continues to see steady growth in its user base. According to user feedback, key metrics for JuiceFS have shown consistent and robust growth, with 2024 being particularly remarkable:
- Total file systems: 30,000+, an 8.5x increase
- Active clients: 107,000+, a 2x increase
- Total files: 300 billion, a 3.3x increase
- Managed data: 700 PiB, a 4.2x increase
In terms of data usage scale, the growth rate in 2024 accelerated significantly. Despite this rapid expansion, the community remained highly active. Over the course of the year, 265 issues were addressed, with a median first response time of 22 hours—indicating that users typically received a reply within a day of raising an issue.
This year, we had a wealth of inspiring case studies across various industries, showcasing the versatility and impact of JuiceFS:
- MiniMax Built a Cost-Effective, High-Performance AI Platform with JuiceFS
- Low-Cost Read/Write Separation: Jerry Builds a Primary-Replica ClickHouse Architecture
- NAVER, Korea's No.1 Search Engine, Chose JuiceFS over Alluxio for AI Storage
- BentoML Reduced LLM Loading Time from 20+ to a Few Minutes with JuiceFS
- vivo Migrated from GlusterFS to a Distributed File System for AI Training
- coScene Chose JuiceFS over Alluxio to Tackle Object Storage Drawbacks
- Hai Robotics Achieved High Availability & Easy Operations in a Hybrid Cloud Architecture with JuiceFS
- TAL: Building a Low-Operation Model Repository Based on JuiceFS in a Multi-Cloud Environment
- How Clobotics Overcame Multi-Cloud and Massive File Storage Challenges
- Metabit Trading Built a Cloud-Based Quantitative Research Platform with JuiceFS
- MemVerge Chose JuiceFS: Small File Writes 5x Faster than s3fs
- iSEE Lab Stores 500M+ Files on JuiceFS Replacing NFS
- Beike Loads AI Models 20x Faster with Hybrid Cloud Storage
- From HPC to AI: Evolution and Performance Evaluation of File Systems
- How Zhihu Ensures Stable Storage for LLM Training in Multi-Cloud Architecture
- From Object Storage to K8s+JuiceFS: 85% Storage Cost Cut, HDFS-Level Performance
Beyond the scenarios mentioned above, JuiceFS has found extensive application in the AI domain, serving a wide range of subfields:
- Generative AI: MiniMax, ChatGLM, StepFun, ModelBest, and more
- Platform applications: NAVER, Xiaohongshu, WPS, Zhihu, LiblibAI, and more
- AI infrastructure: Lepton AI, BentoML, SiliconFlow, Cerebrium, and more
- Self-driving: Momenta, Horizon Robotics, Zhuoyu, and more
JuiceFS is also widely adopted in other AI-powered industries such as quantitative finance, consumer electronics, and biotechnology. We’re deeply honored to have earned the trust of these pioneers in technological innovation.
In 2024, multiple articles about JuiceFS were published by famous technical platforms such as DZone and hit the Hacker News front page. This year also marked JuiceFS' debut on global media platforms like The IT Press Tour and DataTalks.Club.
Gratitude and looking ahead
We sincerely thank every user, contributor, and advocate who has supported JuiceFS throughout 2024. Your engagement and feedback drive our continuous improvement. We especially appreciate those who shared their JuiceFS experiences, inspiring more users to explore its potential.
As we move into 2025, we look forward to helping you achieve even greater efficiency and innovation. Here’s to another year of growth and collaboration!
