A company is developing a mobile gaming app in a single AWS Region. The app runs on multiple Amazon EC2 instances in an Auto Scaling group. The company stores the app data in Amazon DynamoDB. The app communicates by using TCP traffic and UDP traffic between the users and the servers. The application will be used globally. The company wants to ensure the lowest possible latency for all users.Which solution will meet these requirements?
Answer(s): B
Global latency for TCP and UDP worldwide is best served by AWS Global Accelerator with a Network Load Balancer (NLB) endpoints, providing static anycast IPs and TCP/UDP support at scale. A) uses ALB which does not support UDP; B) correctly uses NLB with Global Accelerator integration for TCP/UDP, meeting latency and global reach. C) CloudFront is a CDN for HTTP/HTTPS and does not optimally handle raw UDP/TCP gaming traffic with dynamic backend scaling; D) ALB behind CloudFront does not support UDP and adds unnecessary latency. Therefore B is correct.
A company has an application that processes customer orders. The company hosts the application on an Amazon EC2 instance that saves the orders to an Amazon Aurora database. Occasionally when traffic is high the workload does not process orders fast enough.What should a solutions architect do to write the orders reliably to the database as quickly as possible?
To maximize reliable, fast writes, decoupling via SQS with a scalable worker pool ensures orders are buffered and processed asynchronously, preventing head-of-line blocking during traffic spikes.A) Increasing EC2 size is a scaling step but still relies on a single producer-consumer path; it’s not as reliable as decoupling with a queue for burst traffic. SNS is a pub/sub fan-out not suitable for guaranteed durable processing into Aurora.B) Correct: SQS decouples producers from the database; EC2 Auto Scaling with ALB provides scalable workers to write to Aurora reliably and quickly.C) SNS with direct DB subscription is not durable and is not designed for reliable durable writes to databases.D) CPU-threshold triggering with scheduled scaling is reactive and complex; SQS with a persistent queue and autoscaled workers is more robust.
An IoT company is releasing a mattress that has sensors to collect data about a user’s sleep. The sensors will send data to an Amazon S3 bucket. The sensors collect approximately 2 MB of data every night for each mattress. The company must process and summarize the data for each mattress. The results need to be available as soon as possible. Data processing will require 1 GB of memory and will finish within 30 seconds.Which solution will meet these requirements MOST cost-effectively?
Answer(s): C
Using AWS Lambda with a Python script is most cost-effective for small, fast data processing triggered by S3 uploads, with quick 30-second execution and modest memory needs, leveraging Lambda’s on-demand compute without provisioning clusters.A) AWS Glue with a Scala job is overkill and more costly for small, frequent, rapid jobs; Glue is a managed ETL service with higher startup latency and pricing.B) Amazon EMR with Spark incurs cluster provisioning overhead and higher costs for simple, per-record processing.D) AWS Glue with PySpark also involves Glue overhead and is not as cost-efficient as a short-lived Lambda function for this use case.
A company hosts an online shopping application that stores all orders in an Amazon RDS for PostgreSQL Single-AZ DB instance. Management wants to eliminate single points of failure and has asked a solutions architect to recommend an approach to minimize database downtime without requiring any changes to the application code.Which solution meets these requirements?
Answer(s): A
A) Converting the existing database instance to a Multi-AZ deployment provides synchronous standby in a different AZ, enabling automatic failover with no application code changes, thus minimizing downtime.B) Restoring a new Multi-AZ deployment from a snapshot adds a lag during cutover and requires manual reconfiguration; not seamless for traffic failover.C) Read-only replicas are for load reduction and scaling; failover is not automatic and Route 53-based distribution does not provide automatic failover for write traffic.D) EC2 Auto Scaling with Route 53 does not apply to managed RDS; reconstructing and managing DB instances increases complexity and downtime risk.
A company is developing an application to support customer demands. The company wants to deploy the application on multiple Amazon EC2 Nitro-based instances within the same Availability Zone. The company also wants to give the application the ability to write to multiple block storage volumes in multiple EC2 Nitro-based instances simultaneously to achieve higher application availability.Which solution will meet these requirements?
Provisioned IOPS SSD (io2) with EBS Multi-Attach supports simultaneous writes from multiple EC2 Nitro-based instances in the same AZ, enabling high availability and shared block storage access required by the application. io2 offers high IOPS, high durability, and supports multi-attach for shared volumes across instances.A) gp3 with Multi-Attach is not supported for shared writes across multiple instances in practice; gp3 is general purpose and Multi-Attach is limited. B) st1 is optimized for throughput/hard drive workloads, not suitable for shared block storage across instances. D) gp2 is not guaranteed to support multi-attach in all regions and has lower IOPS compared to io2.
Post your Comments and Discuss Amazon SAA-C03 exam prep with other Community members:
Our website is free, but we have to fight against AI bots and content theft. We're sorry for the inconvenience caused by these security measures. You can access the rest of the SAA-C03 content, but please register or login to continue.