IaaS provides the most control over the underlying infrastructure, including OS and middleware, allowing the organization to install and configure custom logging agents, security information and event management (SIEM) integrations, and have direct access to system-level logs. This level of control is necessary to meet stringent, custom security logging mandates that PaaS or SaaS models might abstract away.

Containers provide operating-system-level virtualization, packaging an application and its dependencies into a standardized unit. They offer stronger isolation than simply running processes on a shared OS (Option D) but are more lightweight and faster to start than full virtual machines (Option B), making them ideal for efficient microservice deployment.

Durability refers to the probability that a stored object will not be lost or become corrupted over a period of time. ‘Eleven nines’ durability indicates an astronomically high likelihood of data survival, meaning it is highly protected against loss. This is distinct from availability (uptime), latency (speed of access), or throughput (data transfer speed).

Right-sizing involves matching instance types and sizes to your workload performance requirements. Since the CPU utilization is consistently low, moving to a smaller, cheaper instance type that can still handle the 15% load is the most direct cost-saving measure. Reserved instances (Option D) save money but lock you into a potentially oversized instance. The other options are more disruptive or not cost-effective for a consistently running, low-utilization workload.

Rapid Elasticity refers to the cloud’s capability to quickly and automatically scale resources outward (scale out) to handle increased load and scale inward (scale in) when the load decreases. This is the ideal characteristic to address predictable traffic spikes, as it allows the e-commerce platform to add resources just before Friday afternoon and release them afterward, ensuring performance and optimizing cost.

The principle of Least Privilege dictates that a user or process should have only the minimum levels of access (permissions) necessary to perform its intended function. By granting the DevOps engineer extensive resource management rights but explicitly denying identity and access management rights, the administrator is directly applying this security best practice.

A managed database service is a classic example of Platform-as-a-Service (PaaS). The cloud provider manages the underlying infrastructure, runtime, database software, patching, and core operations like backups. The consumer (you) manages the data, schema, and typically configures performance-related settings like scaling tiers or read replicas, interacting primarily via connection strings and queries.

To meet a 2-hour RTO, you need a warm or hot standby with relatively current data. Asynchronous replication with a short delay (like 15 minutes) provides a good balance between performance impact (less than synchronous) and data currency (much better than daily backups). This allows the application in the West region to be brought online quickly with minimal data loss (governed by the RPO, not RTO).

This describes a loosely-coupled, multi-tier or microservices architecture. The primary benefit is improved resilience (a failure in one tier doesn’t necessarily crash the entire application) and granular scalability (you can scale the web tier independently of the database tier based on specific load patterns). While it can influence other factors, resilience and scalability are the core direct benefits of this separation of concerns.

IaaS provides the most control over the underlying infrastructure, including OS and middleware, allowing the organization to install and configure custom logging agents, security information and event management (SIEM) integrations, and have direct access to system-level logs. This level of control is necessary to meet stringent, custom security logging mandates that PaaS or SaaS models might abstract away.0