Bringing Browser-Based MFA SSO to the OpenStack CLI
Learn how a lightweight keystoneauth1 plugin brings your existing browser-based MFA and SSO to the OpenStack CLI, with no changes to any client tools.
Production Ceph Since 2015
Enterprise Storage.
Open source. Battle-tested.
Block, object, and file from one cluster — no proprietary licensing, no vendor lock-in. Backed by a team with 10 major Ceph upgrades and zero data loss.
Block (RBD)
Object (S3)
File (CephFS)
100% Data Integrity
100% Upstream Ceph
No fork, no proprietary layers
10 Major Upgrades
Hammer through Squid, zero data loss
Data Sovereignty
Your data center or ours
24/7 Storage Engineers
Ceph specialists, not generalists
Trusted by engineering teams at
Deployed alongside 20,000+ compute cores and 1.7 PB of Ceph storage at the University at Buffalo Center for Computational Research.
Serving production storage workloads across industries
Why Us
Operational Depth You Can't Hire For
The team behind your storage has been running production Ceph for over a decade. That experience shows in every upgrade, every incident, and every architecture decision.
10 Major Upgrades. Zero Data Loss.
Production Ceph since Hammer in 2015 — through every major release to Squid. Every upgrade planned, tested, and executed without losing a byte. Automated zero-downtime migrations.
Upstream Ceph. Open-Source Tooling.
No fork. No proprietary management layer. No per-socket licensing traps. We deploy upstream Ceph with Cephadm and maintain open-source Ansible automation. No vendor lock-in, ever.
Runs Where Your Data Lives
Deploy in our data centers or on your hardware for complete data sovereignty. Encryption at rest, CRUSH-aware failure domains, and compliance-ready architecture for HIPAA and SOC 2.
One Cluster. Three Protocols.
Block storage via RBD, S3-compatible object storage via RGW, and file storage via CephFS. One deployment, one support contract, three ways to access your data — today and tomorrow.
Storage Protocols
Three Protocols. One Cluster. One Team to Call.
Every Ceph deployment includes all three storage interfaces. Use one today, add another tomorrow. No separate products, no extra fees per protocol.
Block Storage
Ceph RADOS Block Device (RBD)
High-performance block volumes for VMs, databases, and Kubernetes persistent volumes. NVMe-backed pools for latency-sensitive workloads. Snapshot-capable, replicated across failure domains, and thin-provisioned. RBD mirroring for DR.
VM disksDatabasesKubernetes PVsNVMe pools
Object Storage
S3-Compatible via RADOS Gateway (RGW)
S3-compatible API for unstructured data at any scale. NVMe-accelerated metadata for high-throughput ingest workloads. Bucket versioning, lifecycle policies, and multi-site replication for geo-distributed durability. Zero egress fees.
Data lakesBackupsMedia assetsNVMe metadata
File Storage
CephFS with NFS Gateway
POSIX-compliant distributed file system for shared workloads. Flash-tier pools for latency-sensitive shared storage. Native CephFS kernel client for Linux, NFS gateway for everything else. Snapshots, quotas, and multi-tenant isolation.
Shared directoriesHPC scratchFlash-tier pools
Use Cases
Storage Problems We've Already Solved
01
Private Cloud Storage Backend
The storage layer underneath OpenStack, Kubernetes, or both. Block volumes for VMs, persistent volumes for containers, and S3 for application data — one Ceph cluster backs the entire stack.
02
AI & ML Data Pipelines
Store training datasets on S3-compatible object storage with no egress fees. Feed data directly to GPU nodes over high-speed networks. Keep raw data, processed features, and model checkpoints in one cluster.
03
Backup & Long-Term Archive
Erasure-coded pools for cost-efficient cold storage. S3 lifecycle policies automate tiering. Store years of backups and compliance archives without per-TB licensing costs.
04
VMware & Legacy Storage Migration
Replace proprietary SAN and NAS with Ceph. Migrate VMware datastores to RBD-backed volumes with MigrateKit. Our phased migration covers assessment, pilot validation, and cutover with no downtime.
Engagement Models
Choose How You Want to Work
Mix and match your infrastructure and engagement model. Deploy on our hardware or yours, with full management or expert support.
Hosted
On-Premise
Fully Managed
Hosted + Managed
We deploy and operate Ceph in our data centers. You consume storage via S3, RBD, or CephFS.
- Zero ops burden
- Predictable per-node pricing
- SLA-backed uptime
On-Premise + Managed
We operate your Ceph cluster in your data center with full 24/7 responsibility for deployments, upgrades, and incident response.
- Full data sovereignty
- Predictable per-node pricing
- Your hardware, our ops
Expert Support
Hosted + Support
Your Ceph cluster runs on our infrastructure with 24/7 expert guidance, architecture reviews, and direct engineer escalation.
- 24/7 Ceph engineers
- Architecture reviews
- Upgrade assistance
On-Premise + Support
You operate the cluster in your facility with direct access to our engineers for troubleshooting, upgrades, and architecture planning.
- Direct engineer access
- Incident escalation
- Upgrade planning
Hosted + Managed
We deploy and operate Ceph in our data centers. You consume storage via S3, RBD, or CephFS.
- Zero ops burden
- Predictable per-node pricing
- SLA-backed uptime
Hosted + Support
Your Ceph cluster runs on our infrastructure with 24/7 expert guidance, architecture reviews, and direct engineer escalation.
- 24/7 Ceph engineers
- Architecture reviews
- Upgrade assistance
On-Premise + Managed
We operate your Ceph cluster in your data center with full 24/7 responsibility for deployments, upgrades, and incident response.
- Full data sovereignty
- Predictable per-node pricing
- Your hardware, our ops
On-Premise + Support
You operate the cluster in your facility with direct access to our engineers for troubleshooting, upgrades, and architecture planning.
- Direct engineer access
- Incident escalation
- Upgrade planning
2015
In production since
First deployed on Ceph Hammer
100%
Data integrity
Across every upgrade and migration
5+ PB
Storage under management
Across hosted and on-premise clusters
10
Major upgrades completed
Hammer through Squid, zero data loss
Evaluate
10 Questions to Ask Any Ceph Vendor
Different organizations approach Ceph support differently. Here's what matters when evaluating providers.
Is it unmodified upstream Ceph, or a vendor fork?
Upstream Ceph, no fork
Does pricing get complicated as you scale?
Simple per-node pricing — no per-socket or per-TB surprises
Is deployment tooling open source or proprietary?
Open-source Cephadm and Ansible automation
Does the vendor run Ceph in production themselves?
5+ PB in production since 2015
How many major Ceph upgrades has the vendor executed?
10 major releases, zero data loss
Does it integrate with Kubernetes and OpenStack?
Native integration via Atmosphere stack
Is the vendor a Ceph Foundation member?
General Member of the Ceph Foundation
Can you leave without a forklift migration?
100% upstream — full portability guaranteed
Are operations and support backed by SLAs?
SLA-backed uptime and response times on all engagement models
Is it cheaper than hiring and retaining a Ceph team in-house?
Predictable per-node costs vs. 3-5 FTE storage engineer salaries
Integration
Fits Your Stack
Ceph works on its own or as the storage layer for your broader infrastructure. Start with what you need. Add more later.
Standalone Ceph
Ceph as a dedicated storage service. S3-compatible object storage, block volumes, or shared file systems without requiring any other platform.
Teams that need scalable storage and already run their own compute.
Ceph + Kubernetes
Persistent volumes backed by Ceph RBD via CSI. S3 object storage for application data. One storage cluster serving all your containerized workloads.
Platform teams running stateful workloads on Kubernetes.
Ceph + OpenStack
Block storage for Nova instances via Cinder. Object storage via Swift-compatible RGW. Image storage for Glance. Ceph is the native storage backend for OpenStack.
Private cloud operators who need unified storage across all OpenStack services.
Full Atmosphere Stack
Ceph, Kubernetes, and OpenStack deployed together as an integrated platform. Storage, compute, and orchestration with one support contract.
Organizations building complete private cloud infrastructure.
Your Storage Runs. Your Team Ships.
Whether you're deploying a new Ceph cluster or migrating off proprietary storage, our engineers have done it before. Book a free 30-minute architecture review — no commitment, no sales pitch.
What you'll get
- Architecture review and capacity planning for your specific workloads — from initial deployment design to growth projections based on a decade of production operations data
- Deployment on your hardware or ours, using upstream Ceph with open-source tooling — no proprietary management layers, no billing surprises after you go live
- Upgrades executed by engineers who have completed 10 major Ceph release transitions — tested against your configuration before touching production
- 24/7 monitoring and incident response from storage specialists who work on Ceph daily — not generalist help desk forwarding tickets
- S3-compatible object storage, block volumes, and CephFS from one cluster with zero egress fees and predictable per-node monthly costs
- Native integration with Kubernetes persistent volumes and OpenStack storage services, or standalone deployment if storage is all you need
Emergency Storage Support
Experiencing a storage incident? Mark your message urgent for immediate 24/7 response from our storage engineering team.
Contact Our Team
Let us help you with your storage needs
Blog
The Latest From Us
Insights, updates, and stories from our team
The Real Cost of Running AI on Hyperscalers vs. Open Infrastructure
Hyperscaler AI looks fast but hides long-term lock-in and rising costs. See how OpenStack and Kubernetes deliver GPU infrastructure you actually control.
Why GPUs Sit Idle: The Hidden Efficiency Problem in AI Infrastructure
Many AI clusters run at only 30–50% GPU utilization. Learn why GPUs sit idle and how Kubernetes, scheduling, and better infrastructure design can improve AI infrastructure efficiency.