VMware Greenplum 7 is now generally available. Greenplum is a unified analytics and artificial intelligence (AI) platform designed to help enterprises make the most of their data resources. The core of VMware Greenplum is based on the open source PostgreSQL project, which is unique in that it seamlessly integrates business intelligence (BI) and artificial intelligence (AI) functions on the same platform.
The announcement states that VMware Greenplum 7 demonstrates its commitment to creating and evolving an inherently secure, mature and flexible SQL-based online analytical processing (OLAP) platform. This innovative platform introduces a series of enhancements and new capabilities focused on resource management and complex analytics capabilities for all data types, whether structured, semi-structured or unstructured. VMware Greenplum 7 also has many important updates in terms of seamless data scalability, multi-workload handling, and deployment flexibility.
New features in VMware Greenplum 7:
- Open Source and PostgreSQL 12 Evolution: VMware Greenplum 7 is built on open source code, leveraging the power, reliability, and flexibility of modern PostgreSQL versions. Compared with the previous version, Greenplum 7 is rooted in PostgreSQL 12 and integrates PostgreSQL releases from the past five years.
- Multiple index types: VMware Greenplum 7 supports multiple index types, including B-tree indexes, hash indexes, bitmap indexes, block range indexes, text indexes, geospatial indexes, and AI vector indexes. This feature optimizes data retrieval and query performance. The Greenplum query optimizer has been continuously improved since 2009, achieving a good performance record in version 6 and expanded in version 7 to provide comprehensive index selection support.
- Enhanced data federation with PXF: The Platform Extensions Framework (PXF) in VMware Greenplum 7 has been improved to enable superior data federation. Enterprises can now query data sets in Amazon Simple Storage Service (S3) object storage, Hadoop Distributed File System (HDFS), and other relational databases via JDBC. It leverages PostgreSQL's foreign data wrapper API to access remote data sources in parallel, providing an abstract data model to manage the security and statistics of remote data to optimize queries.
- Enhanced text search: VMware Greenplum 7 extends text search capabilities with support for both lexical search and AI-driven semantic search to provide more accurate search results. Lexical search supports traditional text search based on keywords, and for semantic search, it is powered by artificial intelligence and vector embeddings.
- Upgraded geospatial analysis: VMware Greenplum 7 has upgraded geospatial analysis capabilities by integrating PostGIS version 3. This improvement greatly increases the speed and feature richness of geospatial queries.
- Row-level security permissions: This feature complements the role-based security model and table-level and column-level permissions already in VMware Greenplum.
- Generated columns for enhanced data modeling: Generated columns were introduced in VMware Greenplum 7, improving data abstraction and modeling, addressing use cases such as security feature-preserving data masking.
- Improved DBA query functionality: Greenplum 7 has made numerous improvements to DBA query functionality, including UPSERT support, user-defined functions with transactions, and improvements to ALTER TABLE to reduce data rewriting.
- Enhanced semi-structured and unstructured data analysis: In addition to supporting XML documents, Greenplum 7 also supports semi-structured data processing, such as enhanced JSON and array data processing capabilities. Full-text search and text-based lexical search indexing enable efficient text storage, indexing, and searching. Additionally, vector embeddings enable condensed and efficient representation of unstructured data, allowing similarity searches for matching documents, images, and videos across multiple languages, including multilingual searches.
- PostgreSQL extension ecosystem: More comprehensive PostgreSQL extension support, such as advanced password checking, fuzzy string matching, Hyperloglog, Ip4r for network data, ISN for media data, nanosecond timestamps, sparse vectors, for pivoting Tablefunc, UUID for unique identifiers, and pg_vector for artificial intelligence vector embedding, are all supported.
- Advanced resource management: Greenplum 7 introduces a series of advanced resource management features. These features ensure robust performance under high load conditions.
- VMware vSphere deployment model: Greenplum 7 can be deployed in bare metal or public cloud environments by referring to the recommended architecture. You can also use the automatic deployment mode provided in Greenplum 7 version to seamlessly integrate into the vSphere private cloud environment.
- Multi-datacenter disaster recovery solution: As part of a multi-datacenter disaster recovery solution, data is replicated through transaction log archives, resulting in more efficient and lower recovery point objectives (RPO) and recovery time objectives ( RTO) disaster recovery solution.
- New extension PostgresML: Provides new user-defined functions that allow users to use tens of thousands of open source artificial intelligence/machine learning pre-trained models in VMware Greenplum.