Total Storage Capacity by Version
Ceph · Production consulting
Make Ceph boring again.
Evidence-first consulting for Ceph clusters in production. I collaborate with your engineers to diagnose issues, reduce risk, and ship changes safely—no guessing.
No-guessing strategy
Observe → form a hypothesis → validate with metrics and controlled changes.
Deliverables
Prioritized plan (P0/P1/P2), runbooks & guardrails, validation checklist.
Engagements
Assessment, incident/performance support, upgrade planning & execution.
I don’t sell “one-size-fits-all” configs or vague advice. If the right solution is “don’t change anything yet—collect evidence first”, that’s what I’ll recommend.
This keeps production safe and avoids expensive “tuning-by-superstition”.
Ceph public telemetry
Snapshots from the public community dashboard.
Version by Cluster Count
Active Clusters (daily, last 7d)
Why Ceph
Unified storage
One system that can serve block, object, and file.
Scale-out economics
Grow capacity and throughput by adding nodes.
Resilience
Designed for failure: recovery and rebalancing are core behaviors.
Vendor flexibility
Open ecosystem, avoids hard lock-in choices.
What I optimize for (no-bullshit)
- Stability under failure and during recovery, not "benchmark-only" tuning.
- Measurable changes with rollback paths.
- Runbooks your team can execute without guesswork.