Examkin

Question

Difficulty: HardPerformance and Scalability Optimization

A national health agency is designing a portal for citizens to retrieve digital health certificates. A nationwide broadcast scheduled for exactly 20:0020:00 UTC will direct citizens to log in and download their certificates. The platform's baseline workload is 1,0001,000 requests per minute, but it is projected to instantly surge to over 800,000800,000 requests per minute within 22 minutes of the broadcast. The backend API is deployed on Amazon ECS on AWS Fargate behind an Application Load Balancer (ALB) and retrieves data from an Amazon Aurora PostgreSQL database. Which architecture strategy should a Solutions Architect recommend to ensure the platform handles the peak traffic spike with minimal latency and no dropped requests?

  1. A
    Submit a support case to AWS to pre-warm the Application Load Balancer to the expected traffic volume. Configure Scheduled Scaling for the Amazon ECS service to scale out tasks before the event. Configure the Amazon Aurora database in a Multi-AZ deployment and configure the application to route read queries directly to the failover standby instance in the secondary Availability Zone.
  2. Submit a support case to AWS to pre-warm the Application Load Balancer to the expected traffic volume of 800,000800,000 requests per minute. Configure Scheduled Scaling for the Amazon ECS service to scale out to the required task count before 20:0020:00 UTC. Configure Aurora Auto Scaling to add Aurora Replicas to handle the read load, and configure the application to use the Aurora reader endpoint.Answer
  3. C
    Configure the Application Load Balancer with a target tracking scaling policy based on the ActiveConnectionCount metric. Implement target tracking scaling for the Amazon ECS service using average CPU utilization. Configure Aurora Auto Scaling to provision additional Aurora Replicas dynamically as read latency increases.
  4. D
    Submit a support case to AWS to pre-warm the Application Load Balancer. Configure ECS Service Auto Scaling based on CPU utilization, setting the scale-out cooldown period to 1010 seconds to accelerate container launch. Scale the primary Amazon Aurora instance vertically to a larger instance size prior to the event, routing all read and write queries to the primary instance.

Answer

The architecture that pre-warms the Application Load Balancer, uses Scheduled Scaling for ECS tasks, and utilizes Aurora Replicas with the reader endpoint.
The correct architecture leverages AWS Support to pre-warm the Application Load Balancer, ensuring it has enough capacity provisioned to handle the massive surge of 800,000800,000 requests per minute from the first second. Scheduled scaling for ECS Fargate ensures that the container tasks are fully booted, healthy, and registered with the target group before the 20:0020:00 UTC start time. Finally, routing read traffic to the Aurora reader endpoint utilizes Aurora Replicas to scale read performance horizontally, preventing database bottlenecks.

Step-by-Step Solution

1
Analyze the traffic pattern and load balancer scaling behavior.
An instant surge from 1,0001,000 to 800,000800,000 requests per minute cannot be handled by the default scaling behavior of the Application Load Balancer, requiring pre-warming.
To ensure the routing infrastructure can handle the immediate load without dropping packets or throttling.
2
Determine the ECS container scaling strategy.
Dynamic target tracking has a bootstrap delay. Scheduled scaling scales out tasks before 20:0020:00 UTC.
To ensure enough compute instances are fully initialized and ready to process requests before the traffic arrives.
3
Optimize the database tier for heavy read operations.
Configure Aurora Auto Scaling with Aurora Replicas and update the application client to query the Reader endpoint.
To distribute the read queries horizontally across replicas instead of overloading the primary writer instance.

Key Concept

Handling massive, scheduled flash traffic spikes in AWS requires pre-allocating capacity at both the routing/load-balancing tier and compute tier, combined with horizontal read scaling at the database tier.
Rate this question

Topics

Performance and Scalability Optimization

More questions from this topic