Database

Cloud SQL

Good for Web frameworks such as CMS, ecommerce.

Cloud SQL is a fully managed relational database service : MySQL, PostgreSQL or Microsoft SQL Server.

Cloud SQL can also scale up, which does require a machine restart or scale out using read replicas.
That being said, if you're concerned about horizontal scalability, you'll want to consider Cloud Spanner.

Performance

• 60TB of storage.
• 60K IOPS.
• 624GB RAM.
• Scale out with read replicas.

Security & Compliance
The data in CloudSQL is automatically encrypted at rest and in transit. External connections can be enforced to be SSL-only.
For secure connectivity you can also use Cloud SQL Proxy which is a tool to help you connect to your Cloud SQL instance from your local machine.

Cloud Spanner

Performance / Good for

Cloud Spanner is especially suited for applications that require :

• SQL relational database management system with joins and secondary indexes.
• Global scale.
• Strong global consistency.
• Built-in high availability.
• Speaks SQL.
• Database sizes that exceed 2 TB.
• And high numbers of input/output operations per second. Tens of thousands of reads/writes per second or more.

Cloud Spanner is a fully managed, mission-critical, relational database service that offers transactional consistency at global scale, automatic, synchronous replication for high availability.

Battle tested by Google’s own mission-critical applications and services.

Firestore

Performance / Good for

• Flexible.
• Scales horizontally.
• NoSQL cloud database.
• Strong consistency guarantees.
• Atomic batch operations.
• Real transaction support.
• Good for Hierarchical, mobile, web such as User profiles/session/preferences, game state, products reviews.

Firestore is a serverless, NoSQL document database.

Firestore offers 2 database modes: Native and Datastore.
When you create a new Firestore database, you must select database mode and you cannot use both Native mode and Datastore mode in the same project.

• Datastore mode should typically be used for server applications.
• Native mode should be used for new mobile and Web apps or when requiring real-time and offline features.

Bigtable

Performance / Good for

• Handle massive workloads.
• Consistent low latency.
• High throughput.
• Good for
  • Heavy read + write, low-latency, events such as Adtech, IoT, financial.
  • Storage engine for ML apps.
  • Operational applications.
  • Analytical applications.
• Easy integration with opensource big data tools(Hadoop, HBase, Spark).

Cloud Bigtable is Google's fully managed Petabyte-scale NoSQL database service for heavy reads/writes.

It's the same database that powers many core Google services, including Search, Analytics, Maps, and Gmail.

The smallest Cloud Bigtable cluster you can create has three nodes and can handle 30,000 operations per second.

How BIG is Bigtable?

Memorystore

Memorystore is a fully managed in-memory data store service for Redis and Memcached.
Memorystore is protected from the internet using VPC networks and private IP and comes with IAM integration to protect your data.

BigQuery

Good for Enterprise data warehouse.

You can store data in BigQuery. But the intended use for BigQuery is Big Data Analysis and interactive SQL querying.

Storage/Database Scalability