High-performance Redis fork with multi-threading and active replication.
## KeyDB MCP Server: Multi-Threaded Redis Fork The **KeyDB MCP Server** integrates KeyDB, the high-performance multi-threaded fork of Redis, into Google Antigravity. This integration provides AI-assisted access to Redis-compatible operations with significantly improved throughput and unique features like active replication. ### Why KeyDB MCP? KeyDB takes Redis to the next level with enterprise features: - **Multi-Threaded**: Utilizes all CPU cores for higher throughput - **Active Replication**: Multi-master replication without conflicts - **FLASH Storage**: Extend memory to NVMe for larger datasets - **Subkey Expires**: Expire individual hash fields - **Full Redis Compatibility**: Drop-in replacement for Redis ### Key Features #### 1. Enhanced Redis Operations ```javascript // Standard Redis commands with better performance await keydb.set("cache:item:123", JSON.stringify(itemData)); await keydb.expire("cache:item:123", 3600); // Subkey expiration (KeyDB-specific) await keydb.expiremember("user:123:tokens", "access_token", 900); await keydb.expiremember("user:123:tokens", "refresh_token", 86400); // Atomic operations across threads await keydb.multi() .incr("stats:pageviews") .lpush("recent:pages", pageId) .exec(); ``` #### 2. Active-Active Replication KeyDB's unique multi-master capability: - Write to any replica, changes propagate automatically - Conflict resolution with last-write-wins semantics - Geographic distribution for global applications - No single point of failure in cluster setup #### 3. FLASH Storage Tier Extend beyond memory limits: - Store less-accessed data on NVMe drives - Automatic hot/cold data tiering - Cost-effective storage for large datasets - Sub-millisecond latency even from FLASH ### Configuration ```json { "mcpServers": { "keydb": { "command": "npx", "args": ["-y", "@anthropic/mcp-keydb"], "env": { "KEYDB_HOST": "localhost", "KEYDB_PORT": "6379", "KEYDB_PASSWORD": "your-secure-password", "KEYDB_THREADS": "4" } } } } ``` ### Use Cases **High-Throughput Caching**: Applications requiring millions of operations per second benefit from KeyDB's multi-threaded architecture. **Global Distribution**: Active-active replication enables writes from any geographic location with automatic synchronization. **Large Dataset Caching**: FLASH storage allows caching datasets larger than available RAM without sacrificing performance. **Token Management**: Subkey expiration simplifies managing multiple tokens or session attributes with different TTLs. The KeyDB MCP Server empowers Antigravity users with enterprise-grade Redis performance and features for demanding workloads.
{
"mcpServers": {
"keydb": {}
}
}