Apache Amoro

Apache Amoro is a lakehouse management system designed to unify batch and streaming data processing across open table formats. It enables infrastructure-independent, lake-native architectures by integrating with compute engines like Flink, Spark, and Trino. Amoro provides a self-managed, pluggable framework that simplifies data warehouse operations and supports multiple table formats including Iceberg, Paimon, Mixed-Iceberg, and Mixed-Hive.

Features

• Support for multiple table formats: Iceberg, Paimon, Mixed-Iceberg, Mixed-Hive
• Self-optimizing engine for compaction, deduplication, and layout optimization
• Unified catalog service compatible with Hive Metastore and AWS Glue
• Rich plugin ecosystem for integration with Flink, Spark, and Kyuubi
• SQL command-line tools and Web UI for management
• Real-time data processing via LogStore with Kafka and Pulsar
• Infrastructure-independent deployment across cloud, hybrid, and private environments

Capabilities

• Manages diverse table formats with unified operations
• Enables stream-batch fusion for real-time and historical data processing
• Provides millisecond-level SLAs for streaming workloads
• Supports schema evolution, table alteration, and metadata management
• Integrates with multiple compute engines for flexible processing
• Facilitates CDC scenarios with efficient streaming reads

Benefits

• Simplifies lakehouse architecture with plug-and-play components
• Reduces storage costs and improves query performance through self-optimization
• Enhances flexibility with support for multiple formats and engines
• Accelerates development and deployment with built-in tools and plugins
• Promotes scalability and adaptability across various infrastructure setups
• Enables unified data governance and catalog management