Isolation between environments

One of the most important questions on a multi-tenant platform is: is my data separated from other customers'? The answer at Dyncloud is yes, by design and across several layers.

flowchart TB
    subgraph PLAT["Platform · Kubernetes"]
        direction TB
        E1["Your environment (isolated)<br/>your application, your data,<br/>your own identity and permissions"]
        E2["Another customer's environment (isolated)"]
        E1 -. "traffic blocked by default<br/>no access between environments" .-> E2
    end

Logical isolation per environment

Each environment runs in its own isolated space within the platform. The resources of one environment (its application, its database, its storage) are not accessible from another environment.

Separate data: each environment has its own database and its own storage. They are not shared between customers.
Separate identity: each environment runs with its own technical identity and its own minimal permissions.
Separate names and paths: each environment is published at its own address.

Network isolation

Traffic between components is segmented. By default, a component can only communicate with what it needs to function; everything else is blocked. This limits the scope of any potential issue to a single environment. More detail in Network security.

Capacity isolation

Each environment has allocated resources (compute and memory) with limits. This prevents the load of one environment from degrading the performance of another ("noisy neighbor").

In summary

Your data lives in a dedicated environment, with its own database, its own identity, its own segmented network and its own resource limits. Isolation is the norm, not the exception.