Cloud Storage allows world-wide storage and retrieval of any amount of data at any time. We dont need to set up auto-scaling ourselves. Cloud Storage autoscaling is managed by GCP. Cloud Storage is an object store so it is suitable for storing photos. Cloud Storage allows world-wide storage and retrieval so cater well to our worldwide audience. Cloud storage provides us lifecycle rules that can be configured to automatically delete objects older than 30 days. This also fits our requirements. Finally, Google Cloud Storage offers several storage classes such as Nearline Storage ($0.01 per GB per Month) Coldline Storage ($0.007 per GB per Month) and Archive Storage ($0.004 per GB per month) which are significantly cheaper than any of the options above.
Ref: https://cloud.google.com/storage/docs
Ref: https://cloud.google.com/storage/pricing

https://cloud.google.com/monitoring/settings/multiple-projects
https://cloud.google.com/monitoring/workspaces

You can use Google Cloud Deployment Manager to create a set of Google Cloud resources and manage them as a unit, called a deployment. For example, if your team's development environment needs two virtual machines (VMs) and a BigQuery database, you can define these resources in a configuration file, and use Deployment Manager to create, change, or delete these resources. You can make the configuration file part of your team's code repository, so that anyone can create the same environment with consistent results.
https://cloud.google.com/deployment-manager/docs/quickstart

The correct answer is C because BigQuery external tables allow BigQuery to query data stored in Cloud Storage without loading it into BigQuery managed storage. This directly satisfies the requirement to minimize storage cost and avoid ingestion overhead. Apache Avro is a supported format for analytical workloads, and partitioning by a relevant column improves query performance and cost control when only subsets of the historical data are needed. Option A loads the data into native BigQuery storage, which increases managed storage usage. Option B is unsuitable because Cloud SQL is not designed for petabyte-scale analytical transaction history or machine learning workloads. Option D also loads data into BigQuery, which violates the requirement to avoid loading overhead.

Comprehensive and Detailed Explanation From Exact Extract:
To adhere to the most secure way and Google-recommended practices, you must avoid the creation and handling of service account key files.
Option B (Impersonation): This method allows the currently logged-in user to temporarily assume the identity and permissions of the service account. It is keyless, meaning no long-lived secret is downloaded or stored on the local machine, drastically improving security. The gcloud config command sets this behavior for all subsequent gcloud commands in the session.
Reference: Google Cloud Documentation - Best practices for using service accounts:
"Avoid creating service account keys... Instead, use service account impersonation or workload identity federation." Reference: Google Cloud Documentation - Service Account Impersonation (Overview):
"With service account impersonation, a user or service can temporarily gain the permissions of a service account... [This] helps you to avoid creating and downloading a service account key."