Reference:
A custom image is a boot disk image that you own and control access to. Use custom images for the following tasks:
Import a virtual disk to Compute Engine from your on-premises environment or from VMs that are running on your local workstation or on another cloud platform. You can manually import boot disk images to Compute Engine, but one disk at a time.

https://cloud.google.com/compute/docs/images

The standby instance is used in high availability to replace the primary instance when failover occurs. The standby instance doesn't appear in the Google Cloud Console. When failover occurs, connections to the primary instance are automatically transferred to the standby instance.
Cloud SQL Key Terms:
Cloud SQL instance
A Cloud SQL instance corresponds to one virtual machine (VM). The VM includes the database instance and accompanying software containers to keep the database instance up and running.
Database instance
A database instance is the set of software and files that operate the databases: MySQL, PostgreSQL or SQL Server.
High availability
Cloud SQL instances using high availability (HA) provide greater reliability than non-HA instances.
HA in Cloud SQL works by having two synchronized instances: a primary instance and a standby instance. Each instance has exactly one VM. Each instance is in a different zone in the same region.
Failover
A failover is when Cloud SQL switches serving from the original primary instance to the standby instance.
Autofailover is a mechanism that automatically triggers failover when a Cloud SQL instance didn't issue a heartbeat in the previous interval.
Standby instances
The standby instance is used in high availability to replace the primary instance when failover occurs. The standby instance doesn't appear in the Google Cloud Console. When failover occurs, connections to the primary instance are automatically transferred to the standby instance.
Clone
When you clone a Cloud SQL instance, you create a new instance that is a copy of the source instance, but is completely independent. After cloning is complete, changes to the source instance are not reflected in the clone, and changes in the clone are not reflected in the source instance.
Replication
Replication is the ability to create copies of a Cloud SQL instance or an on-premises database, and offload work to the copies. The main reason for using replication is to scale the use of data in a database without degrading performance on the primary instance.
Read replica
The read replica is an exact copy of the primary instance. Data and other changes on the primary instance are updated in almost real time on the read replica. Send your write transactions to the primary instance, and your read requests to the read replica. The read replica processes queries, read requests, and analytics traffic, thus reducing the load on the primary instance.
Source server
Replication copies transactions from a primary instance to one or more read replicas. The primary instance is also called the source server. The source server can be a Cloud SQL primary instance, or a server outside of Google Cloud, such as an on-premises server or a server running in a different cloud. If the source server is outside of Google Cloud, we call it Replication from an external server.
Cloud SQL Auth proxy client
The Cloud SQL Auth proxy client is open source software maintained by Cloud SQL. It connects to a companion process, the Cloud SQL Auth proxy server, running on your Cloud SQL instance. You run the Cloud SQL Auth proxy client on your own servers. The Cloud SQL Auth proxy client can be used to establish a secure SSL/TLS connection to the database instance, and/or to avoid having to open the firewall. Authentication is done through Identity and Access Management (IAM).

Reference:
Google Cloud Coldline is a new cold-tier storage for archival data with access frequency of less than once per year. Unlike other cold storage options, Nearline has no delays prior to data access, so now it is the leading solution among competitors.

Store the information in Secret Manager is a secure and convenient storage system for API keys, passwords, certificates, and other sensitive data. Secret Manager provides a central place and single source of truth to manage access, and audit secrets across Google Cloud.
https://cloud.google.com/secret-manager

Google's Data Labeling Service lets you work with human labelers to generate highly accurate labels for a collection of data that you can use in machine learning models.
References:
-> https://cloud.google.com/vertex-ai/docs/datasets/data-labeling-job
-> https://cloud.google.com/ai-platform/data-labeling/docs