Visual overview
Security IN the cloud
Data · identities · permissions · configuration · guest OS where applicable
Security OF the cloud
Facilities · hardware · networking · infrastructure software
Customer identity and data controls complement AWS responsibility for cloud infrastructure.
Technical reference
The boundary moves with service abstraction, but customer ownership of data and configuration does not disappear.
AWS protects facilities, hardware, networking, and managed service infrastructure.
Customers configure identities, permissions, data protection, and workload settings.
Virtual servers leave guest operating-system configuration and patching with the customer.
AWS performs more platform operations while authorization and data governance remain customer concerns.
Security of the cloud and in the cloud
AWS divides cloud security responsibilities between AWS and the customer. AWS is responsible for security of the cloud: the infrastructure that runs AWS services, including the facilities, hardware, software, and networking that make up the AWS Cloud. Customers do not physically secure AWS data centers or maintain the host hardware that AWS uses to provide its services.
The customer is responsible for security in the cloud. Exactly what that includes depends on which services are selected, how those services abstract infrastructure, and how the customer configures them. Customer data, identity and access choices, and many service configurations remain customer concerns even when AWS manages the underlying platform.
How responsibility changes by service
Amazon EC2 is a useful example of a service with substantial customer control. AWS manages the physical infrastructure and virtualization layer, while the customer is responsible for areas such as the guest operating system, including updates and security patches, application software, and the configuration of AWS-provided security controls such as security groups. Greater control over the operating environment brings greater customer responsibility.
More abstracted services move additional platform responsibilities to AWS. AWS uses Amazon S3 and Amazon DynamoDB as examples where AWS operates the infrastructure, operating system, and platform. The customer still remains responsible for its data and for configuration choices such as encryption options, permissions, and identity access. A managed service therefore reduces operational responsibility; it does not transfer ownership of security decisions about customer data.
How to reason about a new service
When encountering an unfamiliar AWS service, ask what layer AWS is operating and what choices remain in the customer's hands. If the customer controls an operating system, that usually brings patching and hardening responsibility. If AWS fully manages the operating system, focus customer attention on data classification, identities, access policies, encryption configuration, network exposure, application behavior, and logging as applicable to that service.
The most common mistake is to memorize one fixed list of responsibilities and apply it everywhere. AWS explicitly states that customer responsibility varies according to the services used, their integration into the customer's environment, and applicable laws and regulations. The model is a boundary for reasoning, not a single universal checklist.
Key takeaways
- 01
AWS owns security of the cloud; customers own security responsibilities in the cloud according to the services they use.
- 02
EC2 exposes more of the stack to the customer than highly managed services such as S3 or DynamoDB.
- 03
Customer responsibility for data, identities, permissions, and configuration remains central.
- 04
Compliance responsibility is shared and must be evaluated for the actual workload.
Official AWS sources
Use these primary AWS resources for the source material behind this article and for deeper reference.