Professional Data EngineerDesigning data processing systemsMedium
A media company needs to process millions of video transcoding jobs daily. Each job involves downloading a source video, applying various transformations, and uploading the processed video to Cloud Storage. The workload is highly spiky, with peak demands occurring during prime-time hours. The company requires a solution that is fully managed, scales automatically, and only charges for the actual compute used. Which Google Cloud product is best suited for this use case?
- AGoogle Kubernetes Engine (GKE) with custom autoscaling.
- BCloud Functions.
- CCompute Engine with managed instance groups.
- DCloud Run.
Show answer & explanationAnswer & explanation
Correct answer: D. Cloud Run.
Cloud Run is a fully managed, serverless platform for containerized applications that scales automatically from zero to millions of requests and only charges for the resources consumed while requests are being processed. This perfectly matches the requirements for a spiky, containerized workload that needs automatic scaling and cost efficiency.
Why the other options are wrong
- A. GKE offers custom autoscaling but requires significant operational overhead for cluster management and is not fully managed in the same serverless sense as Cloud Run.
- B. Cloud Functions are suitable for event-driven, short-lived functions, but video transcoding jobs are typically longer-running and better suited for containerized applications.
- C. Compute Engine with managed instance groups provides autoscaling but requires managing VMs and is not fully serverless, incurring costs even when idle.
Serverless Container Workloads
Running containerized applications in a fully managed, automatically scaling, and pay-per-use serverless environment.
- Ideal for spiky workloads and microservices.
- Scales to zero, meaning no cost when idle.
- Supports any language or library that can be containerized.
Memory trick: Cloud Run: Containers that Scale and Save Cash.