Ross ROSS = Recommend OSS · open-source software intelligence for agents

juicedata/juicefs

JuiceFS is a distributed POSIX file system built on top of Redis and S3. observed · 2026-08-28

github.com/juicedata/juicefs · homepage · Go · Apache-2.0 (permissive) observed · 2026-08-28

Health v2 · maintenance only

98/100

  • Activity 99
  • Release rhythm 96
  • Longevity 100
How is this computed?

round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-02. Adoption (stars, forks) is never an input.

  • gap_med: 19
  • age_days: 2063
  • days_rel: 26
  • days_push: 7
  • n_releases_24m: 14

Full methodology

Adoption not part of the score

14358 stars · 1279 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

JuiceFS is a high-performance, POSIX-compatible distributed file system that stores file data in object storage (e.g., Amazon S3) and metadata in databases such as Redis, MySQL, TiKV, or PostgreSQL. It is designed for cloud-native environments, offering FUSE mounting, an S3 gateway, a Hadoop Java SDK, and a Kubernetes CSI driver.

Use cases

  • mount s3 object storage as a local posix filesystem
  • shared storage for kubernetes pods via csi driver
  • replace hdfs for hadoop big data workloads
  • shared file storage for machine learning training data
  • store billions of files cheaply on object storage
  • multi-cloud and hybrid cloud file storage
  • s3-compatible gateway over a mounted filesystem

When to choose

  • you need a fully POSIX-compatible shared filesystem backed by cheap object storage
  • you want strong consistency across thousands of concurrent clients
  • you run Kubernetes and need persistent shared volumes
  • you want HDFS or S3 protocol compatibility without running a full storage cluster
  • you need high metadata performance with Redis, TiKV, or a database as the metadata engine

When to avoid

  • you need a single-node filesystem with no distributed requirements
  • you cannot operate an external metadata engine like Redis or MySQL
  • you need block storage or raw device access rather than file storage
  • you require fully offline operation with no object storage backend
  • you need a self-contained storage cluster like Ceph that manages its own disks

Facets

application · maturity stable

file-system object-storage caching cli self-hosted microservices cloud-computing big-data infrastructure-as-code files windows go self-hosted cloud fuse posix-filesystem s3-gateway csi-driver hadoop-sdk metadata-engine object-storage-backed containers linux macos docker kubernetes

10 sources

Member repositories

RepositoryRoleHealth v2
juicedata/juicefsmain98

For agents

markdown · JSON · MCP: product_card(name="juicedata/juicefs")

Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem