Professional Cloud Architect · Free Practice Question Hard
Question 22
A media company plans to build a global video streaming platform on GCP. The platform needs to stream content to millions of users worldwide, offering low latency and high availability. The content is primarily video-on-demand (VOD), and the company wants to implement a solution that optimizes the delivery of videos based on the user's location. The platform must also support real-time analytics to monitor viewer engagement and optimize content delivery. Cost efficiency and the ability to scale automatically are critical requirements. Which architecture best meets the company’s requirements?
-
A
Implement the platform using App Engine Standard Environment, Cloud Spanner for metadata storage, and Cloud CDN for video delivery. Store videos in a multi-regional Cloud Storage bucket.
-
B
Use Google Kubernetes Engine (GKE) with multi-regional clusters, Cloud CDN for video delivery, and BigQuery for real-time analytics. Store videos in a single-region Cloud Storage bucket.
-
C
Use Cloud Run for the video streaming application, Firestore for metadata storage, and Cloud Storage with Object Versioning enabled for video storage. Implement Cloud CDN for content delivery.
-
D
Deploy the video streaming application on Compute Engine with a global HTTP(S) Load Balancer and use Cloud SQL for storing metadata. Store videos in a multi-regional Cloud Storage bucket.
Reveal correct answer
Correct answer: A
A.
App Engine Standard Environment offers auto-scaling and high availability, making it suitable for a global streaming platform. Cloud Spanner provides a globally distributed, strongly consistent database, ideal for storing and accessing metadata in real time. Cloud CDN ensures low-latency video delivery by caching content close to users. Storing videos in a multi-regional Cloud Storage bucket further enhances availability and reduces latency, aligning well with the company’s requirements.
B.
GKE with multi-regional clusters and Cloud CDN is a strong combination for global content delivery, but storing videos in a single-region Cloud Storage bucket could result in higher latency for users far from the storage region. Additionally, storing data in a single region introduces a single point of failure, which goes against the high availability requirement. BigQuery is an excellent choice for real-time analytics, but the storage strategy needs improvement.
C.
Cloud Run is serverless and scales automatically, but it may not be the best fit for a high-throughput, latency-sensitive video streaming platform. Firestore is not optimized for large-scale metadata storage and querying, especially under the heavy load of a global user base. Cloud Storage with Object Versioning increases storage costs and complexity without directly benefiting the core requirements of the streaming platform. While Cloud CDN is appropriate for video delivery, the overall architecture is not as robust as the App Engine and Cloud Spanner combination.
D.
While Compute Engine offers flexibility, it does not automatically scale as efficiently as other managed services like GKE or App Engine, which is crucial for handling millions of users. Cloud SQL, while a robust relational database, may struggle with scaling for a global platform, especially under heavy load. Although a multi-regional Cloud Storage bucket is a good choice for video storage, this setup may not provide the best optimization for video delivery in terms of latency.
Discussion
Think the marked answer is wrong, or have a better explanation? Share it below — comments appear after review.
