February 2024 Summaries

PlanetScale has introduced Schema Recommendations, an upgrade to its built-in MySQL database monitoring tool. This feature provides tailored recommendations for improving database performance, reducing memory and storage usage, and optimizing schema based on production database traffic. The Schema Advisor system analyzes query-level telemetry data to generate DDL statements that can be directly applied to a database branch and then deployed to production. Currently, four types of schema recommendations are supported: adding indexes for inefficient queries, removing redundant indexes, preventing primary key ID exhaustion, and dropping unused tables. These recommendations aim to enhance the overall efficiency and performance of databases on PlanetScale.

Foreign key constraints, which enforce referential integrity in databases, are now generally available on PlanetScale. Over 2,400 PlanetScale databases have enabled this feature during the beta phase. To enable foreign key constraints, go to your database's "Settings" page and check the box to Allow foreign key constraints. Foreign key constraints ensure consistency between related data in different tables by enforcing rules on how records can be inserted or deleted. While they can help maintain referential integrity, they may also cause performance issues in high concurrency workloads and add complexity to the database. Developers should carefully consider whether to use foreign key constraints based on their specific application needs.

Amazon Aurora is a scalable database service on AWS that simplifies the process of creating and managing a MySQL or Postgres-compatible database. However, its pricing structure is not as straightforward as one might believe. Factors to consider when estimating costs include instance type, replicas, storage configuration, data transfer costs, cross-region replication, connection pooling, change management, backups, and monitoring. PlanetScale offers a more simplified and managed approach with its pricing model, including some features at no additional cost.

Three common MySQL database design mistakes include suboptimal data type selection, missing or redundant indexes, and improperly storing semi-structured data. Selecting an inappropriate data type can lead to insufficient storage space or over-provisioned columns with performance implications. Missing or redundant indexes can negatively impact query performance and storage costs. Storing semi-structured data as JSON using a dedicated column type can improve efficiency, enable native querying and filtering, and support indexing for faster searches.

PlanetScale has made OAuth applications available for everyone, allowing seamless authentication with the platform and management access to users' organizations and databases. This feature enables developers to build integrations on top of PlanetScale and expand its capabilities across other tools and platforms. Companies such as Netlify, Vantage, Cloudflare, and Seltzer have already built OAuth-based integrations during the limited beta phase. Now, anyone can create an OAuth application in public beta and start building on top of PlanetScale.

PlanetScale is deprecating the Scaler plan in favor of their more popular product, Scaler Pro, which has received positive feedback from customers for its straightforward resource-based pricing. This decision is based on thousands of customers' feedback and a belief that everyone deserves predictable pricing for their database. PlanetScale's architecture was built to handle massive scale applications, making it the home of the best database for serverless. The company believes that being the best database for serverless workloads does not require being serverless, as they focus on holding state for horizontal applications. PlanetScale will provide assistance and recommendations for customers moving from Scaler to Scaler Pro plans.

Amazon Aurora blue/green deployments and PlanetScale branching are two different approaches to deploying and managing database environments. Blue/green deployments involve creating a copy of the existing database environment, making changes to the new environment, and then switchovering traffic to it. In contrast, PlanetScale branching creates isolated MySQL clusters for each branch, allowing developers to test new features without affecting production. Both approaches have their own strengths and weaknesses, including differences in schema change management, instance type adjustments, version upgrades, cost considerations, and potential data inconsistency issues. While blue/green deployments can provide a seamless switchover experience with minimal downtime, PlanetScale's branching approach offers greater flexibility and control over the deployment process, including the ability to roll out changes without disrupting production.