Professional Cloud Architect · Free Practice Question Hard
Question 25
You are a cloud architect for a retail company that has recently migrated its e-commerce platform to Google Cloud. The platform currently handles a peak load of 50,000 simultaneous users during sales events. The company anticipates a 10x increase in traffic over the next three years due to expansion into new markets and expects future traffic spikes to reach up to 500,000 simultaneous users. The company wants to ensure the architecture can scale to meet future demands while minimizing operational overhead. Additionally, the company is exploring the use of AI for personalized recommendations, which could significantly increase the demand on its data processing and storage systems. Which architectural approach should you recommend to ensure scalability and future growth, while also taking advantage of potential cloud and technology improvements?
-
A
Deploy the application in multiple regions and use global load balancing to distribute traffic, paired with autoscaling.
-
B
Use pre-configured VM instances with fixed CPU and memory allocations to handle the expected peak load.
-
C
Migrate the application to a monolithic architecture to simplify management and scaling.
-
D
Implement an auto-scaling group of managed instances with a high upper limit to automatically adjust to varying load.
Reveal correct answer
Correct answer: A
A.
Deploying the application in multiple regions with global load balancing ensures that the traffic is distributed efficiently, reducing latency and preventing any single region from becoming overwhelmed. Paired with autoscaling, this approach can dynamically adjust to both current and future traffic loads. Moreover, multi-region deployment provides resilience and helps in disaster recovery scenarios.
B.
Pre-configured VM instances with fixed CPU and memory allocations might handle the current load but are not flexible enough to accommodate the significant increase in traffic anticipated in the future. Fixed allocations lead to underutilization during non-peak times and might not be sufficient during extreme spikes, requiring manual intervention to scale.
C.
Migrating to a monolithic architecture is a step backward in terms of scalability and flexibility. Monolithic applications are more challenging to scale and manage as they grow, especially under varying and increasing load. Microservices or serverless architectures would be more suitable for handling the anticipated traffic growth and complexity.
D.
While an auto-scaling group of managed instances is a good start, relying solely on this approach without considering other aspects like regional distribution and load balancing can lead to bottlenecks. Additionally, high upper limits might incur unnecessary costs during non-peak periods if not managed carefully.
Discussion
Think the marked answer is wrong, or have a better explanation? Share it below — comments appear after review.
