| Service model | Managed infrastructure and connectivity delivered against a workload-specific responsibility boundary.[1] | EC2 is IaaS under AWS's shared-responsibility model. AWS operates the underlying cloud; the customer manages the guest OS, applications and configurations. AWS Managed Services is a separate enterprise operations service.[3][8] |
| Provisioning | Discovery and design establish dependencies, access paths, backup and ownership before build or migration.[1] | Customers provision through the console, APIs and infrastructure-as-code tooling, choosing instance, image, network, storage and Region. AWS Regions provide multiple Availability Zones for resilient designs.[9] |
| Management responsibility | DC Core can manage guest infrastructure operation, hardening, network controls, monitoring and patching inside scope. Application and data responsibilities are explicitly separated.[1] | For EC2, AWS states that the customer manages the guest OS, updates and patches, application software and security-group configuration.[3] |
| Networking architecture | DC Core can use private addressing, secured ingress and outbound-agent connections through a tenant-aware relay platform.[2] | Amazon VPC gives instances private addresses and optional public addresses. Customers assemble subnets, routing, security groups, load balancers, NAT, endpoints and other controls for their design.[5] |
| Private admin access | Approved agents normally establish outbound platform connections, so inbound SSH/RDP openings are not normally required.[2] | AWS Systems Manager Session Manager also supports node administration without open inbound ports, bastion hosts or SSH-key management. It uses IAM policies for central access control.[6] |
| Backups and DR | Encrypted backup and recovery controls are included where agreed, with retention and service-specific RPO/RTO documented for the workload.[1] | AWS Backup can centralise and automate protection policies across supported AWS and on-premises resources. The customer must configure plans, select resources, monitor jobs and validate that the design meets its recovery need.[7] |
| Location and scale | The standard managed design places production tenancy data in London with disaster-recovery capacity in France, unless another arrangement is agreed.[1] | AWS provides many global Regions. Its public infrastructure guide lists UK, Ireland, France and other European Regions, with at least three Availability Zones per Region.[9] |
| Support | Operational support follows the managed service boundary and can include investigation and remediation work inside that scope. | Basic Support covers account/billing questions, service quotas and self-service resources. Paid plans add technical support and faster response. Platform support should not be assumed to include operating the customer's EC2 guest OS or application.[8] |
| Pricing model | Scoped quote covering selected infrastructure and the agreed management, monitoring, backup and support work. | Component consumption pricing. EC2 supports On-Demand, Savings Plans, Reserved Instances, Spot and Dedicated Hosts; storage, data transfer, public IPv4 and other services can be separate line items.[4][5] |
| Best suited for | Buyers that want a UK-led operator and a defined service around a bounded set of business workloads. | Cloud engineering teams that need global reach, extensive service choice, elastic automation or control over a complex multi-service architecture. |