QuestionQ38

Modernize Infrastructure and Applications with Google Cloud

You are a program manager at a Software as a Service (SaaS) company that provides rendering software to animation studios. Your team requires the ability to schedule scenes as needed and interrupt them at any time so they can be restarted later. Each individual scene render completes in less than 12 hours, and there is no service-level agreement (SLA) for the completion time of all scenes. Results will be stored in a global Cloud Storage bucket. The compute resources are not tied to a single geographic location. This software must run on Google Cloud in a cost-optimized manner.

What should you do?

Explanation

Compute Engine preemptible instances are substantially discounted but can be stopped by Google Cloud, so they are appropriate for fault-tolerant, interruptible batch rendering with no completion SLA. A render duration below 12 hours also fits within the 24-hour maximum runtime for a preemptible VM; progress can be checkpointed to Cloud Storage and resumed after interruption.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!