| Typical use | Smaller websites whose CMS works within a standard web hosting environment. | Smaller websites built on a conventional PHP application and database. | Website projects with several contributors and central project management. | Web applications that need a custom-configured server environment. | Several container services, or applications joining an existing AWS or Kubernetes environment. | Applications integrated into the client’s infrastructure and operating procedures. | Individual web services and APIs running on Workers, including alongside an existing website. |
|---|
| Operational work for your team | Less server administration. The application, content and their maintenance are separate responsibilities. | Less server administration. The CMS, extensions and custom code still need maintenance. | The provider runs the hosting platform. Your applications and containers still need maintenance. | Maintain the guest OS, runtimes, applications and monitoring yourself or through an operations partner. | AWS runs the control plane. Responsibility for nodes varies; applications and permissions remain with the team. | Internal IT and agreed partners maintain the infrastructure and application. | No server administration for Workers. Code, permissions and connected services still need attention. |
|---|
| Technical flexibility | PHP and databases within the plan’s limits. Not a freely configurable server environment. | PHP and MariaDB within the web hosting service. Server configuration is predefined. | Managed applications, with custom containers on suitable plans. Resource limits still apply. | Your choice of operating system and software configuration on virtual servers. | Containers and connected cloud services, within the constraints of the chosen services and cluster policies. | Depends on the available hardware, virtualisation and internal IT requirements. | A platform-specific runtime with execution and resource limits, not a general-purpose virtual server. |
|---|
| Deployment | File transfer and, depending on the plan, SSH. Automation must work with the available access. | File transfer and SSH; check the chosen plan. Set up the deployment process separately. | mStudio, CLI and API. Container deployments can also be automated with mittwald’s tools. | Provision servers through the API. Arrange code or container deployment separately. | Container releases through Kubernetes. CI/CD, approvals and infrastructure management need setting up too. | Through the access methods and release procedures approved by the client’s IT team. | Deploy Worker code with Cloudflare’s tools. CI/CD integration is available. |
|---|
| Growth and traffic spikes | Check resources and plan limits. Upgrade the plan or change the hosting model if needed. | Larger plans are available. More storage will not fix every application bottleneck. | Choose suitable resources and hosting. Limits depend on the plan and application. | Larger or additional servers. Load balancing and data storage need to support the change. | Autoscaling can be configured. Account for databases, quotas and costs as well. | Depends on spare capacity. Additional hardware may require procurement and lead time. | Distributed execution across the platform. Service limits and downstream systems can still be constraints. |
|---|
| Backups and recovery | Check what the chosen plan backs up, how long copies are retained and how restoration works. | Check the plan’s coverage. Plan to restore files and the database together. | Check coverage for each service, explicitly including custom container data and additional services. | Configure server and application backups, including consistent database copies. | Plan backups for each data service. A managed cluster is not a backup of application data. | Use the IT team’s backup arrangements and test recovery of the application. | Version the code. Back up stored data through the relevant services and keep it recoverable. |
|---|
| Cost model | Plan price, contract term and extras. Include the regular price after any introductory offer. | Plan price and additional services. Budget separately for application maintenance. | The hosting service and resources chosen, plus support for your applications. | Resources plus administration. Include storage, backups and networking. | Cluster, compute and additional services. Budget for usage, traffic and operational work together. | Hardware, licences, data centre and staff. Existing capacity is not free. | Plan and usage, plus storage or other services. Account for platform limits and extra consumption. |
|---|
| Moving elsewhere later | Files and databases are usually portable. Check versions, email, DNS and configuration separately. | Conventional applications are usually portable. Check exports and differences in the target environment. | Applications and containers can move; mStudio automation and connected services need adapting. | Common runtimes make moving easier. Infrastructure, networks and data still need to be set up. | Kubernetes helps with moving deployments. AWS services and data may still require substantial changes. | Internal integrations, identity systems and networks can make moving outside the company difficult. | Depends on the use of Workers-specific APIs and data services. Not a drop-in move to a conventional server. |
|---|