Professional Cloud ArchitectDesign and plan a cloud solution architectureMedium
A media company needs to process large volumes of video files (hundreds of terabytes daily) for transcoding, content analysis, and metadata extraction. These jobs are fault-tolerant, can be interrupted, and do not require immediate processing. The company wants to minimize costs while ensuring that the processing capacity can scale up significantly during peak times. Which Google Cloud compute option is most suitable for this workload?
- ACompute Engine custom machine types
- BCloud Functions
- CGoogle Kubernetes Engine (GKE) Autopilot with Spot Pods
- DCloud Run
Show answer & explanationAnswer & explanation
Correct answer: C. Google Kubernetes Engine (GKE) Autopilot with Spot Pods
GKE Autopilot manages the underlying infrastructure, simplifying operations. Spot Pods within GKE provide significant cost savings for fault-tolerant, interruptible workloads like video processing, and Autopilot handles scaling efficiently.
Why the other options are wrong
- A. Compute Engine custom machine types offer flexibility but require manual VM management and don't inherently provide cost savings for interruptible workloads compared to Spot instances/pods.
- B. Cloud Functions are suitable for event-driven, short-lived functions, not typically for large-scale, long-running video processing tasks.
- D. Cloud Run is ideal for stateless, request-driven containers and scales to zero, but for hundreds of terabytes of batch processing, GKE with Spot Pods offers better control and cost efficiency for large, interruptible job queues.
GKE Autopilot with Spot Pods
A fully managed Google Kubernetes Engine mode that automatically scales and manages clusters, combined with Spot Pods for running fault-tolerant, interruptible workloads at a significantly reduced cost.
- Autopilot handles node provisioning, scaling, and upgrades.
- Spot Pods offer up to 90% cost savings compared to on-demand.
- Ideal for batch jobs, video transcoding, and other interruptible tasks.
- Requires applications to be resilient to preemption.
Memory trick: Autopilot spots savings for interruptible media jobs.