Cloud Architecture

A financial services company runs its primary application in a single cloud region. Regulatory requirements mandate a recovery point objective (RPO) of 15 minutes and a recovery time objective (RTO) of 1 hour in the event of a full regional outage. The business wants to minimize ongoing cost while still meeting these targets. Which disaster recovery design best satisfies these requirements?

A retail company runs a stateless web application across an auto scaling group behind a load balancer. During a recent incident, the load balancer continued sending traffic to instances that had crashed at the application layer even though the operating system was still running, resulting in intermittent HTTP 500 errors for customers. The architect wants to ensure that unhealthy instances are automatically detected and replaced without manual intervention. Which design change BEST addresses this requirement?

A media company runs a nightly video transcoding pipeline that processes thousands of independent jobs. The work is fault-tolerant and can be restarted if a worker is terminated, and there is no strict deadline as long as jobs complete before morning. The cloud team wants to minimize compute cost for this batch workload while retaining the ability to scale out to hundreds of workers. Which compute purchasing and scaling strategy best fits these requirements?

A database administrator is migrating a transactional (OLTP) PostgreSQL database to a cloud IaaS virtual machine. The workload generates a high volume of small, random read/write operations and requires consistent sub-millisecond latency to meet application SLAs. The team must select the appropriate storage design for the database data volume. Which storage choice BEST meets these requirements?

A retail company runs its e-commerce platform on cloud VMs. Historical data shows normal traffic requires 20 VMs, but during a 6-week holiday season demand triples and remains sustained at that level for the entire period with little day-to-day variability. The finance team wants to minimize compute cost while guaranteeing capacity is available throughout the peak. Which capacity planning approach best meets these requirements?

A media company hosts a video streaming website on a single region's object storage bucket fronted by a web server. Users in Europe and Asia complain of slow load times and buffering for static assets (images, thumbnails, and pre-recorded video segments), while North American users report acceptable performance. The company wants to reduce latency for global users with minimal changes to its origin infrastructure and predictable cost. Which design change should the architect implement?

A media startup hosts a growing library of on-demand training videos in a single-region object storage bucket. Users worldwide report inconsistent buffering, and the company's monthly egress costs from the storage service have risen sharply as viewership expands. The architecture team must reduce global latency while lowering repeat data-transfer charges from origin storage. Which design change BEST meets both goals?

A DevOps team is migrating a PostgreSQL database into a Kubernetes cluster running on managed cloud infrastructure. The database must retain its data even if the pod is rescheduled to a different worker node after a failure. Which architectural approach should the team implement to meet this requirement?

A company runs a production database server on a public cloud that must operate 24/7 with predictable, steady CPU and memory usage. The workload has run consistently for over a year and is expected to continue unchanged for at least three more years. Finance wants the lowest possible compute cost for this always-on resource without sacrificing availability or accepting the risk of interruption. Which purchasing model should the architect recommend?

A media company stores large volumes of user-uploaded assets in a single-region object storage bucket. Leadership requires that the assets remain accessible for read operations even if an entire cloud region becomes unavailable, and they want a solution that requires minimal ongoing operational effort. Which design approach best meets these requirements?

A gaming company runs a multiplayer platform with active players across North America, Europe, and Asia. Player state (inventory, progress) must be written and read with low latency in each region, and the application requires that writes in any region be acknowledged locally without waiting for a distant primary. The team wants a managed database design that keeps write latency low for all regions simultaneously. Which database design best meets this requirement?

A financial services company runs a legacy transaction-processing system in its on-premises data center that must exchange large volumes of sensitive data with new analytics workloads in a public cloud provider. Compliance requires that this traffic never traverse the public internet, and the analytics team reports that batch transfers must complete within a predictable time window each night. The current site-to-site VPN over the internet suffers from variable latency and throughput. Which connectivity option should the cloud architect recommend?

A financial services company is designing a three-tier application in a single VPC. The public-facing web tier sits in public subnets behind an internet-facing load balancer. The application tier resides in private subnets and must receive traffic only from the web tier, distributed evenly across multiple application instances. Compliance rules prohibit the application tier from being reachable from the internet. Which design best distributes traffic to the application tier while satisfying the requirement?

A development team is refactoring a monolithic application into 30 loosely coupled microservices packaged as containers. The workload experiences variable traffic, requires automated rollouts and rollbacks, self-healing when containers crash, and service discovery between components. The operations team wants to minimize manual intervention as the number of containers grows into the hundreds. Which compute design BEST meets these requirements?

A financial services company is deploying a transactional database in a public cloud. The application requires an RTO of under 60 seconds and cannot tolerate data loss during an availability zone outage. The database must remain writable at all times. Which design best meets these requirements?

A media company runs identical application stacks in two different public cloud providers to avoid vendor lock-in and improve resilience against a full provider outage. Users are globally distributed. The architecture team needs a way to direct users to the healthiest, closest cloud endpoint and automatically stop sending traffic to a provider that becomes unavailable. Which design approach best meets these requirements?

A cloud architect is designing a VPC for a three-tier application. The application servers reside in private subnets and must download OS patches and software updates from public internet repositories. For security, these servers must NOT be reachable from the internet by inbound connections. Which design element should the architect implement to meet both requirements?

A development team is building a mobile gaming backend that must store player profiles, inventory items, and session data. The data structure for each player varies significantly and evolves frequently as new game features are added. The application requires single-digit millisecond read/write latency at massive horizontal scale, with no complex joins or multi-table transactions. Which database design best fits these requirements?

A media company stores 200 TB of finished video projects in cloud object storage. Legal requires that all projects be retained for 7 years, but analytics show that projects are almost never accessed after the first 30 days, and retrieval after that point can tolerate a delay of several hours. The storage bill has become the largest line item in the cloud budget. Which storage design best reduces cost while meeting the retention and access requirements?

A company runs its application tier in VPC-A and a shared analytics service in VPC-B, both in the same cloud region. The teams need the application servers to communicate with the analytics service using private IP addresses, with low latency and no traffic traversing the public internet. There is no requirement for VPC-A to reach any networks that are connected to VPC-B. Which design most efficiently meets this need?