MDDB vs Alternatives

How MDDB compares to other database and content management solutions.

vs Traditional Databases

PostgreSQL / MySQL

Advantages of MDDB:

  • ✅ Specialized for Markdown - Native support, not plain text blobs
  • ✅ Zero Configuration - No database server to install and manage
  • ✅ Built-in Versioning - Automatic revision history without triggers
  • ✅ Simpler Deployment - Single ~29MB binary vs database server
  • ✅ Lower Resource Usage - Minimal memory footprint (~50MB)
  • ✅ Embedded - No network latency, direct file access
  • ✅ Vector Search Built-in - No pgvector extension needed

When to use PostgreSQL/MySQL instead:

  • Need SQL joins and complex relational queries
  • Require multi-table transactions
  • Building traditional CRUD applications
  • Need mature ecosystem (ORMs, GUIs, tools)

vs Document Databases

MongoDB

Advantages of MDDB:

  • ✅ Markdown-First - Purpose-built, not generic JSON storage
  • ✅ Triple Protocol - HTTP, gRPC, GraphQL (MongoDB: wire protocol only)
  • ✅ Smaller Footprint - 29MB Docker image vs ~400MB
  • ✅ Type-Safe APIs - gRPC/Protobuf + GraphQL schema
  • ✅ Simpler Operations - No sharding/replication complexity
  • ✅ Built-in Full-Text Search - No Atlas Search needed

When to use MongoDB instead:

  • Need horizontal sharding at scale (>100GB)
  • Require geo-spatial queries
  • Building complex aggregation pipelines
  • Need change streams for real-time sync

CouchDB

Advantages of MDDB:

  • ✅ Higher Performance - 95x faster writes (29K vs 312 docs/s)
  • ✅ Lower Latency - 34µs vs 3.2ms average
  • ✅ Modern APIs - gRPC + GraphQL vs HTTP-only
  • ✅ Vector Search - Built-in semantic search
  • ✅ Smaller Size - 29MB vs ~200MB Docker image

When to use CouchDB instead:

  • Need master-master replication
  • Require offline-first mobile sync
  • Building distributed systems

vs File-Based Systems

Git + Filesystem

Advantages of MDDB:

  • ✅ Instant Queries - Indexed metadata vs grep/find
  • ✅ API Access - REST/gRPC/GraphQL vs file system calls
  • ✅ Concurrent Access - ACID transactions vs file locks
  • ✅ Rich Metadata - Structured tags vs filename conventions
  • ✅ Better Performance - Optimized for document ops
  • ✅ Vector Search - Semantic similarity impossible with files
  • ✅ Full-Text Search - Built-in inverted index

When to use Git/Filesystem instead:

  • Content authored by non-technical users (GUI editors)
  • Need distributed collaboration (GitHub, GitLab)
  • Require merge conflict resolution
  • Want human-readable diffs

vs CMS Platforms

WordPress

Advantages of MDDB:

  • ✅ Lightweight - 29MB binary vs ~200MB PHP + MySQL
  • ✅ API-First - Built for programmatic access
  • ✅ No PHP/Database - Single Go binary
  • ✅ Version Control - Built-in, not plugin-based
  • ✅ Developer-Friendly - Simple API vs WordPress hooks
  • ✅ Performance - Direct DB access vs WordPress query overhead

When to use WordPress instead:

  • Need visual page builder
  • Require thousands of plugins
  • Non-technical content editors
  • Want mature themes ecosystem

Strapi

Advantages of MDDB:

  • ✅ Simpler - Embedded DB vs separate database server
  • ✅ Smaller - 29MB Docker image vs ~600MB
  • ✅ Faster Startup - Instant vs 30-60s
  • ✅ Purpose-Built - Markdown-specific vs generic headless CMS
  • ✅ Triple Protocol - HTTP + gRPC + GraphQL from start

When to use Strapi instead:

  • Need admin UI for content management
  • Require rich media management (images, videos)
  • Building complex content types with relationships
  • Want plugin marketplace

vs Vector Databases

Dedicated Vector Database Services

Advantages of MDDB:

  • ✅ All-in-One - Vectors + metadata + full-text + storage
  • ✅ Embedded - No separate vector DB service
  • ✅ Markdown-Native - Built for documents, not just vectors
  • ✅ Simpler Architecture - Single database for everything
  • ✅ Lower Cost - No vector DB subscription needed

When to use dedicated vector DB instead:

  • Need >1M vectors at scale
  • Require advanced vector indexes (HNSW, IVF)
  • Building pure semantic search (no metadata/full-text)
  • Need sub-5ms vector query latency

vs Search Engines

Heavyweight Search Engines

Advantages of MDDB:

  • ✅ Zero Configuration - No cluster setup, no index mappings
  • ✅ Embedded - No separate search service
  • ✅ Simpler - Single binary vs Java + heap tuning
  • ✅ Lower Resources - 50MB RAM vs multiple GB for a JVM-based search cluster
  • ✅ Built-in Storage - Documents + search in one place

When to use a heavyweight search engine instead:

  • Need distributed search across nodes
  • Require advanced text analysis (stemming, synonyms)
  • Building log aggregation (ELK stack)
  • Need Kibana dashboards

vs Key-Value Stores

Redis

Advantages of MDDB:

  • ✅ Persistent by Default - No AOF/RDB configuration
  • ✅ Rich Metadata - Structured tags vs string keys
  • ✅ Full Documents - Not just key-value pairs
  • ✅ Revision History - Built-in versioning
  • ✅ Vector Search - Semantic similarity (Redis requires RediSearch)

When to use Redis instead:

  • Need sub-millisecond latency for caching
  • Require pub/sub messaging
  • Building real-time leaderboards
  • Need data structures (lists, sets, sorted sets)

MDDB + Redis: Use MDDB for storage, Redis for caching

vs Static Site Generators

Hugo / Jekyll / Gatsby

Advantages of MDDB:

  • ✅ Dynamic Content - No rebuild needed for updates
  • ✅ API Access - Query content programmatically
  • ✅ Real-time Updates - Instant vs regenerate + deploy
  • ✅ Search Built-in - Metadata + full-text + vector search
  • ✅ Multi-language - Same key, multiple languages

When to use SSG instead:

  • Need maximum performance (pre-rendered HTML)
  • Building purely static sites
  • Want CDN distribution
  • Content changes infrequently

MDDB + SSG: Use MDDB as CMS, SSG pulls content via API for builds

Performance Comparison

Benchmark: 3000 documents, batch inserts

Database Throughput Avg Latency Memory Usage
MDDB (Batch API) 29,810 docs/s 34µs ~50MB
MongoDB 5,176 docs/s 192µs ~200MB
PostgreSQL 4,324 docs/s 231µs ~150MB
MySQL 1,214 docs/s 822µs ~300MB
CouchDB 312 docs/s 3,185µs ~180MB

See also: Performance Benchmarks

Feature Matrix

Feature MDDB MongoDB PostgreSQL WordPress Search Engine*
Markdown Native ✅ ❌ ❌ ❌ ❌
Embedded ✅ ❌ ❌ ❌ ❌
HTTP/JSON API ✅ ❌ ❌ ✅ ✅
gRPC API ✅ ❌ ❌ ❌ ❌
GraphQL API ✅ ❌ ❌ ❌ ❌
Vector Search ✅ ✅ (Atlas) ✅ (pgvector) ❌ ✅ (v8.0+)
Full-Text Search ✅ ✅ (Atlas) ✅ ❌ ✅
Revision History ✅ ❌ ❌ ✅ (plugin) ❌
Document TTL ✅ ✅ ❌ ❌ ✅
Webhooks ✅ ✅ (triggers) ✅ (NOTIFY) ✅ ✅ (watcher)
Multi-language ✅ ❌ ❌ ✅ (plugin) ❌
Auth + RBAC ✅ ✅ ✅ ✅ ✅
Single Binary ✅ ❌ ❌ ❌ ❌
Docker Size 15MB 400MB 200MB 200MB 600MB

When to Choose MDDB

Choose MDDB when you need:

  1. Markdown as primary data format
  2. Embedded database (no separate server)
  3. Version control out of the box
  4. Triple protocol (HTTP + gRPC + GraphQL)
  5. Vector + full-text search built-in
  6. Simple deployment (single binary or Docker)
  7. Low resource usage (<100MB RAM)
  8. Developer-first API design
  9. RAG pipelines with semantic search
  10. Multi-language content management

When to Choose Alternatives

Choose PostgreSQL/MySQL if:

  • Building traditional relational apps
  • Need SQL joins and complex queries
  • Require ACID across multiple tables

Choose MongoDB if:

  • Need horizontal sharding at massive scale
  • Building geo-spatial applications
  • Require change streams

Choose WordPress if:

  • Need visual content editor for non-technical users
  • Want thousands of plugins
  • Building traditional websites

Choose a heavyweight search engine if:

  • Need distributed search cluster
  • Building log aggregation system
  • Require advanced text analysis

Choose dedicated vector DB if:

  • Pure vector search (>1M vectors)
  • Need sub-5ms latency
  • Advanced vector indexing (HNSW, IVF)

← Back to README | See all features →