Alluxio
Data orchestration layer that sits between compute engines and storage systems, caching data for faster, unified access across the two.
- Actively maintained
- Foundation-backed
- Well documented
- Permissive license
Overview
Alluxio sits between compute frameworks (like Spark or Trino) and one or more underlying storage systems (S3, HDFS, and others), presenting a unified namespace and caching frequently accessed data in memory or on fast local disks to speed up analytics workloads without moving the underlying data. It decouples compute from storage location.
Alluxio is a fit for data platform teams running analytics engines against remote or slow storage who want a caching and unification layer, a distinct role from the object-storage systems (MinIO, SeaweedFS) already in this catalog - it sits in front of storage rather than being storage itself.
Why it's in this catalog
Catalog rating: ★★★☆☆ (3/5)
Distinctive data-orchestration/caching layer for analytics workloads, adds real diversity to storage.
- Mature
- Foundation-governed
- Categories
- Storage
- Languages
- Java
- License
- Apache-2.0
Related Projects
CubeFS
Apache-2.0Distributed filesystem and object storage system built for cloud-native, large-scale data-intensive applications.
Actively maintained
- Storage
Velero
Apache-2.0Backup, restore, and migration tool for Kubernetes clusters, protecting both cluster resources and persistent volume data against loss or letting you move workloads between clusters.
Actively maintained
- Storage
Ceph
LGPL-2.1Industry-standard distributed storage system providing object, block, and file storage from a single unified cluster.
Actively maintained
- Storage
Longhorn
Apache-2.0Lightweight, distributed block storage for Kubernetes, designed to be simple to deploy and operate compared to heavier storage systems.
Actively maintained
- Storage
Spotted an error? Suggest an edit on GitHub.