Database Categories & Selection Guide - CodeSwordsman

The most popular open-source relational database and the default for web apps; MariaDB is a compatible fork with nearly identical syntax and protocol

Advanced open-source engine with strong complex queries, GIS via PostGIS and JSONB support

Commercial enterprise database common in finance and telecom core systems; full-featured but expensive to license

Microsoft commercial database tightly integrated with Windows and .NET, also available on Linux

A file-based embedded database with no server process, common in mobile apps and local tools

IBM commercial database widely used in large government and banking core systems

Microsoft desktop database for single-machine data handling; rarely used behind a backend service

In-memory key-value store used for caching, counters, distributed locks and light queues, with several persistence modes

Pure in-memory cache with a simple protocol and high multi-threaded throughput; no persistence or rich data types

The most common document database, storing BSON with a flexible schema for fast-changing product data

Open-source document store with an HTTP/JSON API plus built-in multi-version replication and conflict handling

Highly available, masterless wide-column store suited to massive time-series data and write-heavy workloads

Hadoop-ecosystem wide-column store on HDFS, used for big-data batch and near-real-time storage

The most widely used graph database with the Cypher query language, suited to social graphs and knowledge graphs

Chinese open-source distributed graph database handling hundreds of billions of vertices and edges

Purpose-built time-series database with fast writes and aggregation, suited to metrics and IoT data

The default for cloud-native monitoring, shipping a built-in TSDB and PromQL and usually paired with Grafana

A PostgreSQL extension for time-series data that keeps full SQL support, ideal for existing PG stacks

Chinese high-performance time-series database optimized for devices and sensors with strong write compression

Chinese distributed database compatible with the MySQL protocol, scaling horizontally and covering HTAP workloads

Distributed SQL database compatible with the PostgreSQL protocol, strongly consistent across regions

Google Cloud globally distributed database with strong consistency and automatic sharding, available on GCP

Start with MySQL or PostgreSQL: mature ecosystem, abundant operations talent and cheap managed offerings

Choose PostgreSQL with PostGIS and JSONB for stronger complex queries and indexing

Use Redis for caching, sessions and locks; Memcached fits simple high-throughput caching only

Use Prometheus for collection, and InfluxDB, TimescaleDB or TDengine for long-term storage and aggregation

Pick Cassandra or HBase for write-heavy massive data, balancing throughput against operational cost

Use Neo4j or NebulaGraph for deep relationship traversal instead of recursive joins in a relational database