Research topic
Distributed storage
Distributed storage spreads data across the drives of many servers and serves it as one system. Superintelligence (SI) work puts several kinds of data on it, each written and read in its own way: datasets read in long runs, checkpoints written in bursts and read back to resume, model weights kept by version, and databases and indexes that change in small pieces. Each pool's protection, replication or erasure coding, sets how much raw capacity its data takes and how fast it answers small writes.
Research on distributed storage
The research here works from Ceph's documentation, published papers and vendor specifications: how datasets, checkpoints and model weights are written and read, when a pool suits replication or erasure coding, and how a cluster is sized against the GPU systems that read it. Every figure carries its source and its date.
- Ceph block, object and file storage: which workload uses whichCephFS vs RBD vs RGW, from Ceph's Umbrella 21.2.0 documentation: which workloads need a block device, a shared file system or S3-compatible object storage, and why.Distributed storage7 min readPDF report
- Self-hosted S3-compatible storage: Ceph RGW and the alternativesMinIO alternatives for self-hosted S3-compatible storage, compared from each project's own documentation and license as of 8 November 2026: Ceph RGW, SeaweedFS, RustFS, Garage, Apache Ozone and OpenStack Swift.Distributed storage7 min readPDF report
- Storage for model training: datasets, checkpoints and weightsStorage requirements for model training and checkpointing, from published sizes and measurements: datasets read on every pass, checkpoints written in bursts and weights kept by version.Distributed storage7 min readPDF report
Get new research by email
Work with us on distributed storage
Rillor designs and builds Ceph storage for SI work. Block volumes, object buckets and shared file systems come from one cluster. The cluster sits on servers you own, on storage nodes we supply, or inside a private SI cloud.
About distributed storageNotices
NVIDIA, AMD, Ceph, Kubernetes and related product names are trademarks of their owners and are used only to identify products. No affiliation or endorsement is implied.
Tell us what your storage must hold.
Send what the storage must hold, how fast it grows, how it is read and written, and where it runs.