Google Professional Cloud Developer
Get started today
Ultimate access to all questions.
Your application runs on a managed instance group, and you need to share a large read-only dataset across all instances. The requirements are:
- Each instance must start quickly
- Data must be accessible via the filesystem with minimal latency
- Total solution cost should be minimized
What is the optimal approach?
Your application runs on a managed instance group, and you need to share a large read-only dataset across all instances. The requirements are:
- Each instance must start quickly
- Data must be accessible via the filesystem with minimal latency
- Total solution cost should be minimized
What is the optimal approach?
Exam-Like
Explanation:
The question requires sharing a large read-only dataset across a managed instance group (MIG) with low latency, fast instance startup, and minimal cost.
- Option A (Cloud Storage FUSE) introduces network latency since Cloud Storage is object storage, making it unsuitable for very low latency requirements.
- Option B (copy data via startup script) slows down instance startup due to data transfer time, especially for large datasets, and incurs Cloud Storage egress costs.
- Option C (shared read-only persistent disk) allows multiple instances to attach the same persistent disk in read-only mode. This ensures low latency (block storage), fast startup (no data copying), and minimal cost (single disk storage).
- Option D (snapshot-derived disks) creates multiple disks from a snapshot, increasing storage costs and potentially introducing latency during on-demand data loading from the snapshot.
Option C is optimal: it meets all requirements by leveraging a shared persistent disk for cost efficiency, low latency, and quick instance initialization.