Skip to content

README.md

Latest commit

 

History

History
205 lines (159 loc) · 6.71 KB

README.md

File metadata and controls

205 lines (159 loc) · 6.71 KB
Raw

SharpCoreDB Feature Documentation

Welcome to SharpCoreDB documentation! Here you'll find comprehensive guides for all features.

📚 Feature Guides

Database Features

JOIN Operations with Collation Support

Status: ✅ Production Ready
Highlights:

  • ✅ INNER, LEFT, RIGHT, FULL, CROSS JOINs
  • ✅ Collation-aware string comparisons
  • ✅ Binary, NoCase, RTrim, Unicode collations
  • ✅ Multi-column JOINs
  • ✅ Automatic collation resolution

See Collation Guide for complete documentation.

Collation Support (All Phases)

Status: ✅ Complete (Phases 1-7)

Phase Feature Status
Phase 1 Schema support (CREATE TABLE COLLATE) ✅ Complete
Phase 2 Parser & Storage integration ✅ Complete
Phase 3 WHERE clause filtering ✅ Complete
Phase 4 ORDER BY, GROUP BY, DISTINCT ✅ Complete
Phase 5 Runtime optimization ✅ Complete
Phase 6 Schema migration (ALTER TABLE) ✅ Complete
Phase 7 JOIN operations ✅ Complete

Time-Series Features (Phase 8)

Status: ✅ Production Ready

  • ✅ Gorilla/Delta-of-Delta/XOR codecs
  • ✅ Automatic time-range bucketing
  • ✅ Downsampling to lower resolutions
  • ✅ Retention policies
  • ✅ BRIN-style time-range indexes

🔍 Vector Search & Embeddings

Vector Search Documentation

Status: ✅ Production Ready (v1.1.2+)
Performance: 50-100x faster than SQLite

Highlights:

  • ✅ HNSW indexes for fast similarity search
  • ✅ Multiple distance metrics (Cosine, Euclidean, Dot Product, Hamming)
  • ✅ Scalar & Binary quantization (8-16x memory savings)
  • ✅ Flat indexes for exact search
  • ✅ SQL integration (vec_distance() function)
  • ✅ AES-256-GCM encryption support
  • ✅ Fully async API

Quick Links:

Migration Guide: SQLite Vectors → SharpCoreDB

Status: ✅ Production Ready
Benefits: 50-100x faster search, 5-10x less memory, 12-30x faster index builds

Highlights:

  • ✅ 9-step migration process
  • ✅ Data migration strategies
  • ✅ Query translation (SQLite → SharpCoreDB)
  • ✅ Index tuning and optimization
  • ✅ Performance benchmarking
  • ✅ Troubleshooting guide

For You If:

  • ✅ Currently using SQLite vector extensions
  • ✅ Want 50-100x performance improvement
  • ✅ Need native .NET vector support
  • ✅ Scaling vector search workloads

Quick Links:

🚀 Performance Highlights

Vector Search Performance

Operation SharpCoreDB SQLite Speedup
k-NN search (1M vectors) 2ms 100ms 50x
Index build (1M vectors) 5s 60s 12x
Memory (1M vectors) 1.2GB 6GB 5x less

JOIN Performance

JOIN Type Collation Overhead
Hash JOIN (Binary) Baseline
Hash JOIN (NoCase) +1-2%
Nested Loop (NoCase) +5-10%

Aggregate Performance

  • SIMD Analytics: 682x faster than SQLite
  • SIMD Analytics: 28,660x faster than LiteDB

🆕 Recently Completed Features

Phase 1.3: Stored Procedures & Views

Status: ✅ Complete

  • CREATE PROCEDURE with IN/OUT/INOUT parameters
  • CREATE VIEW and CREATE MATERIALIZED VIEW
  • EXEC command with parameter binding

Phase 1.4: Triggers

Status: ✅ Complete

  • BEFORE/AFTER triggers
  • INSERT/UPDATE/DELETE events
  • NEW/OLD column binding
  • Trigger body execution

📖 Documentation by Use Case

New to SharpCoreDB?

  1. Start here: Main README
  2. Learn the API: Manual
  3. Understand features: This page

Using Vector Search?

  1. Quick start: Vector README
  2. Configuration: Vector Configuration
  3. Migration: From SQLite
  4. Performance: Tuning Guide

Using JOINs & Collations?

  1. How it works: Collation Guide
  2. Examples: SQL Examples
  3. Rules: Collation Resolution

Migrating from Another Database?

  1. From SQLite: Vectors Migration
  2. Storage format: Migration Guide
  3. Performance tuning: Benchmark results

🎓 Learning Path

Beginner (New Users)

  1. Main README overview
  2. Quick Start examples
  3. Basic CRUD operations
  4. First SQL query

Intermediate (Developers)

  1. Vector search basics
  2. JOIN operations
  3. Collation support
  4. Performance tuning

Advanced (Architects)

  1. Vector migration strategies
  2. Time-series compression
  3. Index tuning
  4. Scaling to production

📊 Feature Matrix

Feature Status Version Performance Security
Vector Search (HNSW) ✅ Complete 1.1.2+ 50-100x SQLite AES-256 ✅
Vector Quantization ✅ Complete 1.1.2+ 8-16x memory ✅
JOIN Collations ✅ Complete 1.1.2 +1-10% ✅
Time-Series ✅ Complete 1.1.1+ 80% compression ✅
Stored Procedures ✅ Complete 1.1.0+ N/A ✅
Views & Materialized Views ✅ Complete 1.1.0+ 2-100x ✅
Triggers ✅ Complete 1.1.0+ 0-5% ✅

Need Help?

Last Updated: January 28, 2025
All Features: Production Ready
Build Status: ✅ Passing