TangoDB v1.0.0 - High-Performance Embedded Key-Value Storage

TangoDB v1.0.0

TangoDB v1.0.0 is a high-performance embedded key-value storage engine built from scratch in modern Java.

Maven Central Dependency (pom.xml)
<dependency>
    <groupId>io.github.gokul-ceo</groupId>
    <artifactId>tangodb</artifactId>
    <version>1.0.0</version>
</dependency>

TangoDB v1.0.0 Quick Example
// Initialize configuration with FFM native off-heap Arena & WAL enabled
TangoConfig config = TangoConfig.builder()
        .sstableDirectory(Path.of("data"))
        .memTableSize(64L * 1024 * 1024) // 64 MiB MemTable
        .build();

// Open database instance (automatically replays WAL on startup)
try (TangoDB db = TangoDB.open(config)) {
    db.put(bytes("user:1001"), bytes("{"name":"Alice"}"));
    
    byte[] value = db.get(bytes("user:1001"));
    System.out.println("Retrieved: " + new String(value));
    
    db.delete(bytes("user:1001"));
}

⚡

40.77M+ Ops/Sec MemTable Ingest

Designed for extreme throughput, achieving 40,769,881 ops/sec in JMH single-threaded off-heap MemTable record write benchmarks (~24.5 ns latency).

🔍

Bloom Filter & Sparse Indexing

Combines an embedded Bloom Filter (1% false-positive rate) for instant O(1) rejection of missing keys with a Sparse Index for O(log N) binary search range navigation, increasing disk point read throughput by +34% (up to 512.5 ops/sec).

🛡️

WAL Startup Crash Recovery

Automated startup crash recovery with CRC32C binary checksum validation, sequence restoration, delete tombstones, and obsolete log cleanup.

🚀

Java FFM & Off-Heap Memory

Uses JDK 21+ Foreign Function & Memory (FFM) API (MemorySegment and Arena) to store binary record payloads natively off-heap, eliminating JVM Garbage Collection pressure.