Office Hours #12 – JuiceFS Tiered Storage: Hot ⇄ Cold in One Command

​Object storage is inexpensive in general, but costs can ramp up very quickly when you store PB-scale cold or archive data in the standard tier. Don't let it burn your budget until it's too late.

​Starting with JuiceFS v1.4, tiered storage lets you map files or directories to different object storage classes (Standard, Infrequent Access, Glacier, etc.) directly through the metadata configuration. In this office hour session, we'll show you how to keep hot data on fast storage while transparently moving cold data to archive classes—saving costs without changing your application code.

​We'll cover configuration, applying tiers recursively, rewriting objects after mapping changes, restoring archives, and best practices for all kinds of workloads.

​Agenda

• ​17:00 – 17:10 PST | Object Storage Classes Explained

• ​17:10 – 17:25 PST | The Basics: Configuring Storage Tiers in JuiceFS

• ​17:25 – 17:35 PST | Advanced Usage: Rewriting Objects, Restoring Archives

• ​17:35 – 17:45 PST | Live Q&A – Ask the JuiceFS Team Anything

​Speaker: Joe Zhou, Juicedata Developer Advocate

​

​JuiceFS  Office Hours

​The JuiceFS Office Hours is a recurring online event held on the fourth Thursday of each month. It serves as a dedicated platform for JuiceFS users and community members to:

• ​Share our practices and community experiences in the AI field

• ​Receive immediate responses to their questions from the JuiceFS team

• ​Learn about the latest product updates and feature releases

​

​About JuiceFS

​JuiceFS is an open-source, high-performance distributed file system for the cloud. With full POSIX compatibility, it turns object storage into massive local disks across platforms and regions, and is widely used in AI applications like autonomous driving, generative AI, quant research, and biotech.