dabooDb

DabooDb is a pure Java, key-value store, designed for high performance on small to moderate-sized databases. It's named after my daughter, who thought her name was "Daboo" when she was little - thanks to her big brother.

This software was originally written for use in a course on Java performance engineering, but the course never materialized and so the slower "starter" version was jettisoned and this higher-performance version is all that remains. It does work, however ;).

The API

The database provides standard key-value operations:

• Write(), Get(), and Delete(); for both single documents and batches
• GetRange(); for range queries

The Features

What makes dabooDb different?

• It's persistent; Durable in the event of a shutdown or crash
• Supports in-process (embedded) and client-server modes
• Transactions on batch operations
• Optimistic Locking, so it can be safely used by more than one client
• Integrated document compression and encryption
• Integrated document validation using JSR 349 annotations before writes
• Built-in backup and restore
• Extremely flexible: Select or create plugins for:
  • Storage technology and in-memory data structures
  • Communications
  • Compression
  • Encryption
• These options can be combined to create a custom storage solution for your application
• It is quite fast for both reads and writes:
  • 1 million random inserts: 8.8 seconds (113,636 OPS or 8.8 micros per insert)
  • 1 million random reads: 3.279 seconds (304,971 OPS or 3.3 micros per read)

Getting Started

Download the latest release from:

https://github.com/lwhite1/daboodb/releases

Build and install:

mvn clean install

Then add a dependency to your pom file:

<dependency>
    <groupId>org.daboodb</groupId>
    <artifactId>daboo</artifactId>
    <version>1.0-SNAPSHOT</version>
</dependency>

A (rough) roadmap:

1.1

• Pluggable comparators
• Optimistic locking
• Transaction support

1.2

• Performance testing framework
• Performance logging

2.0

• High-Performance client-Server mode with new communication plugin

2.5

• Document schema version management

Beyond that?

Here are some further extensions to make it a full database (if that would be fun):

3.0

• Secondary indexes
• A sql-inspired query language
• Reasonable performance on new query functionality with moderate scale

4.0

• Catalog

5.0

• Full-text-search support