Keynote: Building and Sustaining Momentum in Open-Source Storage
summary
Company experience selling Ceph hardware and running real clusters shows the core problem: device density outpaces cluster recovery and IO plumbing. Tests report Seagate HAMR 30 TB disks on single SATA links and vendors talking about 256 TB NVMe parts, which create rebuild and throughput bottlenecks. Work focuses on tighter device alignment, end-to-end performance tuning, and reducing Ceph overhead on NVMe; Crimson is expected to help and teams chase billion IOPS experiments while pushing doc and operational improvements.
AI-written summary · about
chapters (8)
- 0:00Introduction and welcome
- 1:15Who we are at 45 Drives
- 2:21Contributing to Ceph
- 3:44Success with Ceph for everyone
- 7:36Challenges and competition
- 9:20Denser storage media demand
- 11:05Focus areas for improvement
- 13:52Closing and call to action
related talks
13mKeynote: Getting to Know Ceph
29mCeph and the future of Open Source Storage
08mKeynote: A Wonderful Journey On Ceph
29mCeph's journey at SUSE
01h 02mIntroduction to Ceph and Architectural Overview
35mOptimising resource allocation in hyperconverged infrastructure
01h 50mKeynote
29mCeph, Xen, and CloudStack: Semper Melior