Ceph Object Storage Roadmap Session at Cephalocon 2025
summary
Squid adds object headers with replication status for workflow checks, per-bucket S3 logging with a synchronous journal mode for incremental backups, RGW IAM accounts (IM API compatible) for tenant self‑service, STS/JWT fixes and hybrid-cloud archival that can both push and retrieve objects via restore or transparent read‑through. Tentacle work targets scale bucket indexes to 10 billion with non‑pausing re‑shard, Prometheus per‑user/bucket metrics, per‑storage‑class quotas and stats, whole‑object dedup, NFS over RDMA experiments, and a D4N write‑back cache plus S3 vectors + Lance DB integration.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 0:22Squid release features
- 4:25RGW accounts
- 5:02Account management workflow
- 5:59Data archiving and retrieval
- 7:38Tentacle release overview
- 10:21Bucket index scaling
- 10:35Security and policy updates
- 11:18Performance and deduplication
- 15:33High-speed AI data access
- 16:12Future development plans
- 22:29Advanced tenant security
- 24:11Analytics foundations
- 26:43S3 vectors and AI integration
- 28:26D4N distributed caching
- 30:22Community Q&A
- 31:48Scaling and final discussion
related talks
09mStorage: managing billions of objects
10mKeynote: Ceph as Part of the Data Infrastructure for Zoned Storage
24mAvoiding the Curious Case of Leaky Cauldrons
21mWhat's new in Amazon S3
22mKeynote: gRPC, A Foundation for AI
43mFuture-proofing your applications with AWS databases
35mExtending Kubernetes: Our Journey & Roadmap
33mThe advent of the open data lake