Ceph Days Seattle 2025
Bringing Ceph to Seattle ¶
A full-day event dedicated to sharing Ceph's transformative power and fostering the vibrant Ceph community with the community in Seattle!
The expert Ceph team, Ceph's customers and partners, and the Ceph community join forces to discuss things like the status of the Ceph project, recent Ceph project improvements and roadmap, and Ceph community news. The day ends with a networking reception.
Important Dates ¶
- CFP Opens: 2025-03-30
- CFP Closes:
2025-04-25Extended to 2025-05-02 - Speakers receive confirmation of acceptance: 2025-05-10
- Schedule Announcement: 2025-05-10
- Event Date: 2025-05-15
Schedule ¶
| What | Who | When |
|---|---|---|
| Welcome, Check-in, Coffee, Pastries | 12:00 - 13:00 | |
| Introduction to Ceph: The State of the Cephalopod in 2025 New to Ceph? Or a seasoned operator curious about the latest updates? This talk is your fast track to understanding Ceph in 2025. We’ll cover what Ceph is, how it works, and where the project is headed — from new features and architectural changes to project governance and ecosystem growth. Whether you're deploying your first cluster or managing petabytes, this session will bring you up to speed. | Dan van der Ster (CLYSO) | 13:00 - 13:30 |
| Choosing the Right Data Protection Strategies For Your Ceph Deployments Choosing the right data protection strategy for Ceph deployments can be complicated: Usable to raw capacity ratio Replication vs erasure coding EC profile values for k and m Read and Write performance Recovery performance Failure domains Fault tolerance Media saturation min_alloc_size vs IU Moreover, a given Ceph cluster often benefits from or even needs a combination of strategies and media types. These decisions can be daunting, and many clusters require or would benefit from a mixture of approaches based on use-case and per-pool requirements. | Anthony D'Atri (IBM) | 13:30 - 14:00 |
| Ceph Solution Design Tool In this talk we'll be going over general cluster design recommendations and how to employ those using the Ceph Solution Design utilities. We'll also discuss some of the tradeoffs between using EC vs replicas, sizing of the ratio between HDD and flash in hybrid configurations, and more | Steven Umbehocker (OSNEXUS) | 14:00 - 14:30 |
| Coffee / Tea break | 14:30 - 15:00 | |
| Ceph in Proxmox VE Proxmox embraced Ceph early on, and now this has become very relevant for those migrating virtualization platforms. This talk will provide a technical overview of Ceph implementation in Proxmox VE, Proxmox Backup server, and ISS experiences. | Alex Gorbachev (ISS) | 15:00 - 15:30 |
| Ceph Durability: How Safe Is My Data? How many nines does your cluster really have? Whether you’re running a small 50-disk setup or a hyperscale 5000 OSD deployment, understanding Ceph’s actual data durability is key to making the right design choices. Replication vs. erasure coding, failure domains, recovery speeds: these all impact real-world reliability. In this talk, we introduce a new Ceph durability calculator based on Monte Carlo simulations to give updated, practical insights into how safe your data really is with Ceph. Bring your cluster size and settings — and walk away with numbers. | Dan van der Ster (CLYSO) | 15:30 - 16:00 |
| Optimizing Ceph RGW for Specific Workloads Including AI One of the big barriers to getting great performance and scalability in object storage configurations has to do with data placement. Inefficiently writing small objects to EC storage can kill performance and cause unintentional wasted space due to padding. We'll talk about Ceph's RGW's support for embedded Lua and how we've used that in QuantaStor to solve these issues. We'll then dive into Lua examples you can customize to optimize your object storage workloads. | Steven Umbehocker (OSNEXUS) | 16:00 - 16:30 |
| NVMe over TCP and Block Performance A seasoned IT professional with over 20 years of leadership experience in technology solutions and consulting, Mike specializes in data center modernization, cloud architectures, and disaster recovery strategies. Currently serving as a Technical Product Manager for IBM Storage Ceph, he focuses on NVMe over TCP and VMware vSphere integration for block storage. His expertise spans decades of IT Consulting, public peaking, customer education, strategic planning and high-value solutions architecture. | Mike Burkhart | 16:30 - 17:00 |
| Dinner & Drinks & Networking oh my! | 17:30 - 19:00 |