We use Google Cloud SQL for relational database usage. For storing the user information, and several logical relations on the user entity like grouping of user. We use it only for storing the data, and it already covers the maintenance, so storing and retrieving data is our only use cases
Pros
Uptime
Less latency
Easy to maintain
Cons
Better UI
Easier way of connecting to it for debugging
Likelihood to Recommend
It is very well suited when the team wants low maintenance from their end and they need a solid relation database with all the SQL related features. It is very fast and scaled instantly based on the need and it has the redundancy, so data will be not lost.
It is not well suited, when the team wants full control over the SQL server.
VU
Verified User
Engineer in Engineering (Computer Hardware company, 201-500 employees)
For certain aspects of the BI Landscape, we required the data to be hosted in the cloud rather than on-premises. Architecture. - A sidecar database is deployed in Google Cloud SQL as a supplement to the main on-premise data warehouse. - Data is replicated from on-prem systems and sensor data - Cloud SQL acts as a read-optimized layer for BI tools and sharing to other cloud services.
Pros
Performance scaling during peak use.
Operational Simplicity.
Avoids licensing costs of expensive middleware to expose on-premise systems to external cloud services.
Cons
More customer control wrt. backup schedule and retention.
More granular segmentation for workload limits depending on which pipeline accesses the data to avoid costly errors or misuse.
Enhanced remote API for stored procedures, enabling initiation from on-premise databases to streamline orchestration and monitoring.
Likelihood to Recommend
Does what it promises well, for instance, as a sidecar for the main enterprise data warehouse. However, I would not recommend using it as the main data warehouse, particularly due to the heavy business logic, as other dedicated tools are more suitable for ensuring scalable operations in terms of change management and multi-developer adjustments.
VU
Verified User
Employee in Corporate (Food & Beverages company, 1001-5000 employees)
Our product, [...], which a powerful marketing intelligence tool designed specifically for automotive dealerships, runs on Google Cloud. We're leveraging instance-based services, such as Compute Engine, Google Cloud SQL, Cloud Memory, as well as netsec services, such as Load Balancer and Firewalls. We're getting a robust and centralized solution, that provides us with comprehensive metrics on the consumption for each resource as well as monitoring tools to proactively keep our infrastructure up and running.
Pros
High-availability
Dynamic resource allocation
Serverless
Monitoring tools
Cons
Access through VPN
Defragmentation tools
Automatic version upgrade
Likelihood to Recommend
For lean IT teams with stable data structure Google Cloud SQL is the ideal solution. If a larger team has complex data structures that require constant tuning and maintenance, as well as control over server resources it wouldn't be best suited.
We migrated from using on-prem SQL databases to Google's managed MySQL database, CloudSQL, a few years back. It supports all of our relational database needs. Along with other Google-managed data structures, Cloud SQL plays an integral role for our software team and our R&D teams, as it is used as the primary storage for company-wide analysis.
Pros
It is fully managed, so it gives the user a hands-off approach to relational storage.
It has well-built APIs, so it is easy to access in a multitude of ways.
It is reasonably easy to use with the Google Cloud Console, so users can access it using a user interface.
Cons
There is some functionality you cannot do from the console, including, granting and restricting user privileges, among other database management capabilities.
It is not trivial to connect to spun up VMS. Currently, we use the Cloud SQL proxy to do so.
Likelihood to Recommend
Overall, Cloud SQL serves all of the needs we need it to. It can do everything a relational database can do, plus it is completely managed so you don't have to worry about the size of the VM, partitioning, etc.
