What problem does it solve?
This Skill solves the problem of efficiently ingesting data into Databricks Delta tables with high-throughput and durability, eliminating the need for message buses and enabling direct record-by-record ingestion.
Core Features & Use Cases
- Direct Ingestion: Ingest data directly into Delta tables without a message bus.
- High Throughput: Support for high-throughput data ingestion up to 100 MB/s.
- Durability: Provides durability guarantees with at-least-once delivery semantics.
- Multi-Language Support: SDKs available for Python, Java, Go, TypeScript/Node.js, and Rust.
- Schema Generation: Generate Protobuf schemas from Unity Catalog table definitions.
- Use Case: Ideal for building producers that write directly to Unity Catalog tables, especially in data pipelines that require real-time data ingestion.
Quick Start
Install the Zerobus Ingest SDK for your target language and configure the environment variables. Use the SDK to create a stream, ingest records, and handle acknowledgments.