Scale without rebuilding your infrastructure
Add instances, tune resources, or let autoscaling do the work, all from the same Git, CLI, Console, or API workflow you already use.
The problem
Static infrastructure can't keep up. Most teams hit the same three walls:
Fixed sizing fails under load
Spikes overwhelm what's deployed, and the rest of the time it sits idle.
Overprovisioning is expensive
Sizing for the worst case all year round costs more than it should.
Autoscaling stopped at the app tier
Most platforms autoscale apps but route database scaling to a third-party engine, a Kubernetes operator, or a manual upgrade workflow. Capacity at the database tier never caught up.
Scaling for every workload
Scaling and performance on Upsun allow teams to adjust application capacity through the console, CLI, or API.
- Horizontal and vertical scaling for apps, workers, and managed PostgreSQL and MariaDB
- Native autoscaling across applications, background workers, and managed database read replicas
- No third-party database engine or Kubernetes operator to manage
- CPU and memory-based autoscaling
- Configurable thresholds and instance limits per environment
- Container profiles for any workload type
- Guaranteed CPU for resource-intensive workloads
- Manage from Git, CLI, Console, or API
How scaling works on Upsun
Add or remove instances on demand
Run more copies of your app or worker in seconds. New instances are added without downtime and start serving traffic immediately. Trigger it from the CLI, Console, or API.
Tune resources vertically
Increase CPU, memory, or storage for your app, databases, background jobs, and other services. Choose from four built-in sizes that cover most types of work.
Set CPU or memory thresholds
Upsun adds and removes capacity automatically across apps, workers, and managed PostgreSQL or MariaDB read replicas, with no restart or downtime.
Scale workers independently
Worker processes scale separately from your application. Add worker instances to handle background jobs and queues without touching your app's capacity.
Scaling built around your business growth
Reliability and availability
Every app and service runs in its own container with allocated CPU, memory, and storage. Guaranteed CPU reserves dedicated compute for resource-intensive workloads.
Cost control
Resource allocation can be tuned per environment to balance performance and cost, especially for non-production workloads.
Operational simplicity
Apps, workers, and managed databases share one autoscaling primitive, one platform, and one bill. The same controls apply across the stack, with the integration managed by the platform.
Capacity matched to demand
Set autoscaling rules before peak periods. Capacity grows when load rises and shrinks when it falls, across apps, workers, and managed database read replicas.
Per-environment resource control
Run preview and staging at smaller profiles, reserve larger ones for production. Non-production work never carries production costs.
Growth without migration
As your application grows, add more CPU, memory, and instances on the same platform. No infrastructure migration or application rewrites needed.
Scale your application without the complexity
Join our monthly newsletter
Compliant and validated