Pool many disks into one elastic store — file, object, and block storage for containers, big data, and AI.

Welcome to CubeFS: Cloud-Native Distributed Storage.

CubeFS is a cloud-native distributed storage system that pools the disks of many machines into one elastic, fault-tolerant store accessible as a file system, object storage, or block storage. This free book teaches it from the ground up: distributed-storage fundamentals and trade-offs, CubeFS's architecture (the master, metadata, and data subsystems), data layout with partitions and extents and protection via replication and erasure coding, the POSIX/S3/block access interfaces, Kubernetes integration through CSI including shared ReadWriteMany volumes, multi-tenancy with volumes and quotas, performance tuning across the metadata and data paths, reliability and durability through redundancy and automatic recovery, and operating a cluster in production. Ten focused chapters with clear diagrams that show how a fleet of disks becomes dependable, scalable, multi-protocol storage for modern workloads.

This title is part of the ShriIra library and is free to read in full, right here — our small contribution to making world-class knowledge easy to reach.

A note on reading it: open the Contents menu at the top of the reader to jump between chapters, use the Aa menu to set a comfortable text size, theme (light, sepia, or night), and single- or two-page layout. Your place is saved automatically, so you can always pick up where you left off.

We hope it serves you well.

— Shriira Press

Contents

1. Chapter 1 — What CubeFS Is
2. Chapter 2 — Distributed Storage Fundamentals
3. Chapter 3 — Architecture
4. Chapter 4 — Data Layout and Protection
5. Chapter 5 — Access Interfaces: File, Object, and Block
6. Chapter 6 — CubeFS and Kubernetes (CSI)
7. Chapter 7 — Multi-Tenancy and Management
8. Chapter 8 — Performance and Tuning
9. Chapter 9 — Reliability and Durability
10. Chapter 10 — Operating CubeFS and Putting It Together