Not a black box. Not a shared PaaS. Real VMs in your VPC with the ops burden removed.
Most cloud providers run managed services on shared pools. Multi-tenant clusters where your database shares resources with hundreds of other customers. It's cheaper to operate. But it comes with real problems:
Someone else's query spike affects your latency.
Performance varies based on cluster load, not your workload.
A vulnerability in one tenant's config can expose the whole pool.
You can't tune settings because you don't own the instance.
These invariants apply to every managed service. PostgreSQL, MySQL, Redis. Same model.
Your database isn't a shared tenant. It's a real VM running in your account, with the compute class you choose.
Same private network, same firewall rules, same visibility. No special tunnels or hidden endpoints.
Value, Standard, Premium. Same IOPS guarantees as regular volumes. You pick the performance tier.
Baseline, Standard, Edge. Same bandwidth limits and traffic packages as your VMs.
No magic. Same performance model, same pricing primitives. Predictable behavior across all services.
Great tent production problem beef culture put. Be not as submit say version managing where. Moving say based win minimize.
All managed services share the same architecture. Different software on top.
Same VM lifecycle
Same storage engine
Same network isolation
Same monitoring & backups
Same CLI & API
Quick guide based on your use case
Most versatile. JSON, full-text search, extensions.
Recommended:
Web apps, SaaS, APIs
From ~€10 /mo
In-memory cache. Sub-millisecond latency.
Recommended:
Sessions, cache, queues
From ~€5 /mo
Classic RDBMS. WordPress, legacy apps.
Recommended:
CMS, legacy, PHP apps
From ~€10 /mo
Document store. Flexible schema, JSON-native.
Recommended:
Node.js, flexible data
From ~€10 /mo
Typical setups for different environments
Affordable, can burst when needed. Good for CI/CD, staging, local dev.
~€10-15 /mo
Dedicated resources, consistent performance. 10K IOPS for fast queries.
~€50-60 /mo
Max memory for large datasets. 50K IOPS for analytics/OLTP.
~€150-200 /mo
Typical setups for different environments
Clear division of responsibility. You own the infrastructure, we handle the ops.
Your database runs on a dedicated VM. Not a multi-tenant cluster with noisy neighbors.
You choose the resources. You see the VM. You SSH in when needed.
VM cost + storage cost + management fee. Same primitives, same rates, clear math.
Standard PostgreSQL, standard MySQL, standard Redis. No proprietary extensions.
Managed services are part of the same platform. Same CLI, same API.
$ scamp managed create postgres \ --name prod-db \ --version 16 \ --vm-class dedicated-m \ --storage-class standard \ --storage-size 100 \ --network-class standard \ --vpc prod-vpc Creating managed PostgreSQL 'prod-db'... VM: dedicated-m (2 vCPU, 4 GB RAM) Storage: 100 GB standard (10,000 IOPS) Network: standard (1 Gbit, 20 TB included) VPC: prod-vpc Status: provisioning $ scamp managed list NAME TYPE VERSION STATUS VM CLASS IP prod-db postgres 16 running dedicated-m 10.0.1.50 cache redis 7.2 running burst-s 10.0.1.51 $ scamp managed connect prod-db psql (16.1) Type "help" for help. prod-db=#
Launch a managed service in minutes. Full control, zero ops.
No credit card required.
