The fundamentals of the Managed File System offering
The service Managed File System is a fully managed distributed file system deployed on Cloud Temple's sovereign SecNumCloud infrastructure. Based on reference distributed storage technologies (CephFS), it offers shared access to data via standard protocols. NFS v4.1 and SMB 3.x, and native integration with Kubernetes via the CSI driver.
Designed for application workloads requiring shared file space, this service eliminates the complexity of deploying and maintaining a NAS or distributed file cluster. Cloud Temple manages the entire lifecycle of the platform, from OSD nodes to protocol gateways.
The architecture is based on a cluster extended over several availability zones (Multi-AZ), ensuring that a major outage in one datacentre does not result in any loss of data or interruption of service on the production clusters.
Our compliance procedures
Our Managed File System offering is HDS and ISO 27 001 certified
The benefits of Cloud Temple's Managed File System offering
Simplified operation
Zero Ops, business focus
Entrust complete management of the file cluster (OSD, MDS, gateways, updates) so that your teams can concentrate solely on exploiting the data.
Service continuity
Multi-datacentre resilience
Ensure data availability with triple replication and automatic access failover, even in the event of site failure.
Uninterrupted flexibility
Hot compatibility and scalability
Facilitate migration without application modification (NFS/SMB) and adapt capacity in real time, while supporting modern uses via Kubernetes.
Sovereignty and security
Total control of your data
Guaranteed residency in France on a SecNumCloud infrastructure, with complete network isolation and no public exposure of endpoints.
The key features of our Managed File System
Protocol NFS v4.1
Mount file volumes via NFS version 4.1 from any Linux/Unix or Windows system. Lift & Shift migration without application modification.
SMB 3.x protocol
Access to shares via SMB 3.x (Windows, macOS, Linux compatible). Support for Active Directory authentication (Kerberos).
Kubernetes integration (CSI)
Dynamic provisioning of Persistent Volumes via the CephFS CSI driver. Support for ReadWriteOnce (RWO) and ReadWriteMany (RWX) modes.
Quotas by volume
Application of quotas (maximum capacity) at sub-volume/directory level to control consumption by project or application.
Hot resize
Extension of the capacity of a volume without interruption of service or reassembly.
Snapshots
Creation of consistent snapshots of subvolumes for point-in-time backup and restore.
Transit encryption
TLS 1.3 encryption for NFS communications (via Kerberos/TLS) and SMB 3.x (native SMB Encryption).
Encryption at rest
AES-256 encryption of data stored on OSDs.
Observability (Prometheus/Grafana)
Metrics exposed via ceph-mgr's Prometheus module. Pre-configured Grafana dashboards (IOPS, throughput, latency, capacity, OSD status).
Audit logs
Access and operation logs sent to the customer's centralised log sinks (Syslog, S3). Exhaustive traceability for compliance audits.
Isolation by dedicated cluster
Each customer has its own CephFS cluster on a dedicated infrastructure, with an isolated private network.
Technical specifications
Do you have a Kubernetes file migration or shared storage project? Let's talk.
Do you want to replace an ageing NAS (Lift & Shift), deploy secure Windows-based collaborative shares (SMB/Active Directory) or provide shared persistent volumes (ReadWriteMany) for your Kubernetes pods? Our experts will help you size your cluster, choose the right protocols (NFS v4.1, SMB 3.x, CSI) and configure total network isolation on our SecNumCloud infrastructure.
Tell us about the specifics of your project using this form: we'll get back to you as soon as possible to design the distributed file architecture best suited to your needs.
Pricing
| Units of work Managed file systems - (excluding infrastructure) | 0 units | Unit price € excl. tax/month | Commitment |
|---|---|---|---|
| MANAGE FILE SYSTEM - PRODUCTION (including 3 Nodes) | 1 virtual cluster | 1 260,00 € | 1 month |
| MANAGE FILE SYSTEM - DEV/TEST/RECETTE (including 3 Nodes) | 1 virtual cluster | 675,00 € | 1 month |
| Additional NODE for Managed File Service - PRODUCTION | 1 additional broker node | 262,50 € | 1 month |
| Active Directory module | 1 module | 225,00 € | 1 month |
Use cases
No, the migration is transparent (Lift & Shift).
The service natively supports industry-standard protocols: NFS v4.1 (Linux/Unix, macOS, ESXi) and SMB 3.x (Windows, macOS). You don't need to do any refactoring: only the mount target changes to your new Cloud Temple private endpoint.
No, the network is totally isolated.
Storage endpoints (VIP NFS-Ganesha and SMB Samba) are exclusively exposed on your own private networks (via IPsec VPN or dedicated interconnection). No public IP address is assigned to the storage gateways, drastically reducing your attack surface.
Yes, ReadWriteMany (RWX) mode is fully supported.
The Managed File System integrates natively with your Kubernetes cluster via the official CephFS CSI driver. This enables dynamic provisioning of Persistent Volumes capable of being read and written simultaneously by several pods (e.g. distributed web applications, IA/ML, CMS).
Thanks to “Soft Delete” and Snapshots.
The service includes a Soft Delete function (a recycle bin configurable by subvolume) available on both NFS and SMB. This feature allows you to keep deleted files for a configurable length of time before they are finally purged. You can also create consistent snapshots for point-in-time restores.
Recovery is continuous and is carried out using your standard tools.
In compliance with the Data Act, your files remain your exclusive property and are not linked to any proprietary format. Reversibility is self-service and continuous via native copy tools from your NFS or SMB mounts (such as rsync or robocopy). If you cancel, we guarantee secure erasure (purging) of your data within 7 days.