Examkin

Question

Difficulty: MediumPerformance and Scalability Optimization

A national education board is hosting an online portal for high school exam results. The portal experiences virtually no traffic throughout the year, but when results are released, it experiences an immediate influx of 80,00080,000 requests per second within a 55-minute window. The current architecture consists of an Application Load Balancer (ALB) distributing traffic to Amazon EC2 instances in an Auto Scaling Group, with a backend Amazon RDS for MySQL Multi-AZ database. During a mock release, the system failed due to ALB HTTP 503 (Service Unavailable) errors and database exhaustion from read queries. How should the solutions architect redesign the architecture to handle the next release?

  1. A
    Enable read traffic routing to the standby instance of the RDS MySQL Multi-AZ deployment to distribute the read query load. Rely on the default auto-scaling behavior of the Application Load Balancer to scale out as client connections increase.
  2. B
    Deploy Amazon ElastiCache for Memcached to cache database query results, configuring multi-AZ replication to ensure the cache layer is highly available. Rely on target tracking scaling policies on the Application Load Balancer to automatically scale capacity during the event.
  3. Configure Amazon CloudFront in front of the Application Load Balancer to cache static assets. Migrate the backend database to Amazon Aurora MySQL, and configure Aurora Replicas with Auto Scaling to scale read capacity. Contact AWS Support to pre-warm the Application Load Balancer before the results release.Answer
  4. D
    Migrate the database to Amazon Aurora MySQL. Rely on the Application Load Balancer's automatic scaling to handle the incoming request rate, and configure the EC2 Auto Scaling group with a very short cooldown period to launch new instances as quickly as possible during the spike.

Answer

Configure Amazon CloudFront in front of the Application Load Balancer to cache static assets, migrate the database to Amazon Aurora MySQL with Auto Scaling Aurora Replicas to handle dynamic reads, and request AWS Support to pre-warm the ALB prior to the release.
The correct solution involves offloading static content caching to Amazon CloudFront to reduce load on the origin, migrating the database to Amazon Aurora MySQL to scale reads horizontally via Aurora Replicas, and pre-warming the Application Load Balancer by contacting AWS Support. Pre-warming ensures the ALB is provisioned with sufficient capacity to handle the sudden, massive traffic spike immediately without dropping connections.

Step-by-Step Solution

1
Identify the cause of the HTTP 503 errors.
The ALB was unable to scale quickly enough to handle the instantaneous 80,00080,000 requests per second spike.
ALBs scale based on traffic patterns and need time to scale out. Pre-warming is required for sudden, massive spikes.
2
Address the database read exhaustion.
Migrate to Amazon Aurora MySQL and enable Aurora Replicas with Auto Scaling.
Aurora Replicas scale reads horizontally to handle massive query volumes, whereas RDS Multi-AZ standbys cannot serve read traffic.
3
Optimize asset delivery and reduce load on the ALB.
Deploy Amazon CloudFront in front of the ALB.
CloudFront caches static content closer to users, reducing the volume of requests hitting the ALB and EC2 instances.

Key Concept

Handling massive, instantaneous traffic spikes by pre-warming load balancers, offloading static content using CloudFront, and scaling database reads horizontally with Aurora Replicas.
Estimated Time:2m 0s
Rate this question

Topics

Performance and Scalability Optimization

More questions from this topic