The release of the distributed database Apache Cloudberry 2.1.0 has been announced, continuing the development of the open codebase of the Greenplum database, which was turned into a proprietary product by Broadcom after acquiring VMware. The project is still in the Apache incubator and will be transitioned to a top-level Apache project once the infrastructure and support are ready.
The Cloudberry DBMS is a distributed edition of the open-source PostgreSQL DBMS, optimized for executing analytical queries over large datasets (Data Warehouse). It utilizes a massively parallel processing (MPP) architecture for parallel data processing, ensuring the scalability of the repository to petabyte sizes by segmenting data and employing a cluster of nodes for their storage and processing. servers.
Among the improvements in Apache Cloudberry 2.1.0:
- The UDP2 protocol has been implemented for interaction between nodes, increasing the efficiency of distributed query execution.
- Support for the MCP server (Model Context Protocol) has been added to simplify integration with tools based on large language models.
- The ability to use the LZ4 algorithm for compressing table columns has been added to reduce I/O and memory consumption.
- The performance of the ORCA optimizer has been improved.
- The script greenplum_path.sh, used for setting up the database user's environment, has been replaced with cloudberry-env.sh.
- The backup tools have been renamed to cloudberry-backup. A plugin for S3 protocol-based storage has been included in the main repository.
Source: opennet.ru
