AWS → OVHcloud

Migrating from AWS to OVHcloud

OVHcloud is Europe’s largest cloud provider (French, HQ Roubaix) and carries the broadest managed catalogue of the sovereign set, including managed Kafka, Cassandra and OpenSearch that most EU providers lack. That makes a streaming- or data-heavy AWS estate move cleaner here. This page is specific about where OVHcloud is strong, the two AWS services with no twin, and the DR lesson from the 2021 Strasbourg fire.

Which AWS services have a true OVHcloud equivalent?

Graded conservatively, row by row. OVHcloud’s managed Kafka is the standout that separates it from a Scaleway-shaped migration.

AWS services and their OVHcloud equivalents, each with a migration verdict.
AWSOVHcloudVerdict
Amazon S3OVHcloud Object Storage (S3) S3-compatible with three storage classes (High Performance, Standard, Infrequent Access) and no egress fees. Legacy Swift endpoints also exist, so target the S3 API.Equivalent
Amazon EBSOVHcloud Block Storage Network block volumes with snapshots. Direct equivalent.Equivalent
Amazon EC2OVHcloud Public Cloud Instances General-purpose, GPU and dedicated instances across France, Germany, UK and Poland.Equivalent
Amazon EKSOVHcloud Managed Kubernetes (MKS) Managed control plane integrating instances, GPUs, load balancer, registry and managed databases.Equivalent
Amazon MSK / KinesisOVHcloud Public Cloud Databases for Kafka Managed Apache Kafka (with Connect and MirrorMaker) is a real differentiator; the Kinesis API is not replicated, so target Kafka.Partial
Amazon RDS / AuroraOVHcloud Managed Databases (PostgreSQL) Managed PostgreSQL, MySQL, Cassandra, Valkey and OpenSearch; Aurora-specific features have no twin.Partial
Amazon ElastiCacheOVHcloud Managed Databases for Valkey Managed Redis/Valkey; in-memory caching is well covered.Equivalent
Amazon CloudFront / Lambda@EdgeOVHcloud CDN Content delivery is covered; edge-compute primitives (Lambda@Edge-equivalent) are thin.Partial
AWS LambdaOVHcloud Functions HTTP and cron work; Lambda’s native event-source breadth is not matched.Partial
Amazon DynamoDBNo managed twin (self-host ScyllaDB / Cassandra) No managed serverless wide-column NoSQL with pay-per-request billing.No managed twin
Amazon RedshiftNo managed twin (self-host ClickHouse) No sovereign serverless warehouse; ClickHouse is a different operating model.No managed twin

Where do the APIs actually diverge?

OVHcloud closes more of the managed catalogue than its peers, so the gaps are narrower, but they are still specific.

  • Managed Kafka exists (a genuine edge over Scaleway), but the Kinesis API itself is not replicated. Kinesis consumers retarget to Kafka clients.
  • Object Storage is S3-compatible with no egress fees, though older buckets may be Swift-based; standardise on the S3 endpoints.
  • Edge compute (Lambda@Edge/CloudFront Functions equivalent) is thin. Global edge logic is re-architected, not lifted.
  • DynamoDB and Redshift have no managed twin and drive the self-hosting effort.

Which of your services move cleanly?

Scoped to the rows OVHcloud covers as managed services. Search or filter by verdict.

10 capabilities

  • Object storage

    Hyperscaler Amazon S3 Azure Blob Storage Google Cloud Storage
    EU-sovereign path Scaleway Object StorageOVHcloud Object Storage
    Verdict Mature

    S3-compatible API. A drop-in target for most SDKs, backups and static assets.

  • Block storage

    Hyperscaler Amazon EBS Azure Managed Disks Persistent Disk / Hyperdisk
    EU-sovereign path Scaleway Block StorageOVHcloud Block Storage
    Verdict Mature

    Network block volumes with snapshots. Standard building block, well covered.

  • Virtual machines

    Hyperscaler Amazon EC2 Azure Virtual Machines Google Compute Engine
    EU-sovereign path Scaleway InstancesOVHcloud Public CloudHetzner Cloud
    Verdict Mature

    General-purpose and dedicated instances are a solved problem in the EU.

  • Managed Kubernetes

    Hyperscaler Amazon EKS Azure AKS Google Kubernetes Engine (GKE)
    EU-sovereign path Scaleway KapsuleOVHcloud Managed Kubernetes
    Verdict Mature

    CNCF-conformant managed control planes. Portable workloads move with little friction.

  • Container registry

    Hyperscaler Amazon ECR Azure Container Registry Google Artifact Registry
    EU-sovereign path Scaleway Container RegistryOVHcloud Managed Private RegistryHarbor (self-hosted)
    Verdict Mature

    OCI registries are commodity. Self-hosted Harbor is a fully sovereign fallback.

  • Managed PostgreSQL

    Hyperscaler Amazon RDS / Aurora Azure Database for PostgreSQL Cloud SQL / AlloyDB for PostgreSQL
    EU-sovereign path Scaleway Managed DatabaseOVHcloud Managed DatabasesAiven for PostgreSQL
    Verdict Viable

    Standard PostgreSQL is well served. Aurora-specific features (e.g. global database, serverless v2 autoscaling) have no exact twin, so plan around them.

  • Authoritative DNS

    Hyperscaler Amazon Route 53 Azure DNS Google Cloud DNS
    EU-sovereign path Scaleway Domains & DNSOVHcloud DNSBunny DNS
    Verdict Viable

    Managed authoritative DNS is well covered. Route 53's routing policies need a manual rebuild.

  • Serverless functions (FaaS)

    Hyperscaler AWS Lambda Azure Functions Cloud Run functions (Cloud Functions)
    EU-sovereign path Scaleway Serverless FunctionsOVHcloud Functions
    Verdict Partial

    The runtimes exist, but Lambda's breadth of native event sources and mature cold-start tuning is not matched. Fine for HTTP and cron; harder for deep event-driven fan-out.

  • Global CDN & edge compute

    Hyperscaler Amazon CloudFront / Lambda@Edge Azure Front Door Cloud CDN / Media CDN
    EU-sovereign path Bunny.netGcoreOVHcloud CDN
    Verdict Partial

    EU-headquartered CDNs deliver content well globally. Edge-compute primitives (equiv. to Lambda@Edge/CloudFront Functions) are thinner and less mature.

  • Managed ML platform

    Hyperscaler Amazon SageMaker Azure Machine Learning Vertex AI
    EU-sovereign path OVHcloud AI Endpoints / AI TrainingScaleway GPU Instances + Managed Inference
    Verdict Gap

    GPU compute, training and inference endpoints exist, but nothing matches SageMaker's end-to-end breadth (pipelines, feature store, model registry, tuning) as one managed platform. You assemble it.

Verdicts are conservative and reflect managed, EU-jurisdiction offerings as of 2026. Provider feature sets move quickly, so we re-check on every engagement.

How many engineer-days does this take?

30–75 engineer-days

A directional range for a mid-sized estate. A streaming-heavy estate lands lower here than on providers without managed Kafka.

  • A mid-sized estate: ~20–40 VMs, managed PostgreSQL, object storage, Kubernetes, and a Kafka or streaming workload.
  • The low end assumes S3-, Kubernetes-, PostgreSQL- and Kafka-shaped workloads OVHcloud covers as managed services.
  • The high end assumes a DynamoDB or Redshift dependency plus cross-region DR hardening.
  • Excludes application rewrites, egress time and ramp-up on the OVHcloud API and console.

Where does OVHcloud actually run?

A wider EU footprint than most sovereign providers: France, Germany, the UK and Poland for Public Cloud, plus non-EU regions you can simply not select.

  • Gravelines (GRA), FranceFlagship Public Cloud region with multiple OpenStack regions.
  • Strasbourg (SBG), FranceRebuilt and hardened after the 2021 fire; design DR across regions.
  • Roubaix (RBX), FranceHistoric core site.
  • Frankfurt (DE), GermanyGerman data-residency option.
  • London (UK)UK region (post-Brexit adequacy applies).
  • Warsaw (WAW), PolandCentral-Europe footprint.

What are the honest limitations?

  • The 2021 Strasbourg (SBG) fire In March 2021 a fire destroyed the SBG2 datacentre and some customers without off-site backups lost data. OVHcloud rebuilt and hardened the site, but it is the reason to design cross-region backup and DR explicitly rather than assume it.
  • More do-it-yourself than AWS The console, API and support tiers expect you to lean on your own runbooks rather than white-glove support; budget for that operational load.
  • No DynamoDB or serverless-warehouse twin Those retarget to self-hosted ScyllaDB and ClickHouse, the two rows that survive OVHcloud’s broad managed catalogue.

What else does a CTO ask before committing?

Does OVHcloud’s managed Kafka really change the migration?
Yes, for streaming estates. Providers without managed Kafka push you to self-host or use Aiven; OVHcloud runs Kafka (with Connect and MirrorMaker) as a managed service, so an MSK-based pipeline moves closer to configuration. The Kinesis API itself still has no twin.
How do we avoid another SBG-style data loss?
Treat any single datacentre as fallible: replicate object storage across regions, keep database backups off-site, and validate restore drills. That is standard practice on AWS too. The 2021 fire just makes it non-negotiable here.
Is OVHcloud sovereign for our compliance posture?
OVHcloud is French with EU regions and no US parent, clearing most GDPR and residency requirements. The UK region falls under post-Brexit adequacy, so keep regulated data in France or Germany if that matters.