You are discussing scaling requirements with a gaming company. When the game launches, they are expecting incoming data surges of 2 million users or more during weekends and holidays. Their on-premise

Question

Options

• AEither Compute Engine VMs or Kubernetes nodes work, but it is better to keep a buffer of an
• BWe can deploy a Pub/Sub to ingest data which will grow to absorb demand and pass it on to
• CWe will allocate Compute Engine VMs estimating 80% capacity of 2 million users.
• DWe will allocate Kubernetes nodes estimating 80% capacity of 2 million users.

• A
  5% (3)
• B
  82% (54)
• C
  3% (2)
• D
  11% (7)

Explanation

Cloud Pub/Sub is a fully managed, serverless messaging service that automatically scales to absorb massive, unpredictable data ingestion surges without pre-provisioning capacity. It acts as a buffer between the high-volume incoming data and downstream processing systems, decoupling producers from consumers. This means the system gracefully handles sudden spikes of 2 million+ users without dropping data or crashing. Options A, C, and D all involve pre-allocating fixed compute capacity (VMs or Kubernetes nodes), which requires guessing peak load and risks under-provisioning or wasteful over-provisioning - exactly the on-premise problem the client already faces.

No community discussion yet for this question.