Analytical database PostgreSQL version (formerly HybridDB for PostgreSQL) to provide you with simple, fast and cost-effective cloud PB level data warehousing solutions. Analytical database PostgreSQL open source version is compatible with Greenplum data warehouse, as a fully parallel architecture of MPP number of warehouse services using its broad compatibility with PostgreSQL / Oracle syntax ecology, a new generation of vector engine performance beyond traditional database engine more than 10 times, distributed SQL implement complex query optimizer free tuning. Can be achieved for ad hoc query and analysis of massive data analysis database PostgreSQL version, ETL processing and visualization exploration, the industry is competitive on cloud data warehouse solutions.
The main function
- Easy to fit, free tuning.
Support for SQL 2003, part of the Oracle compatible syntax, supported by PL / SQL stored procedures. A new generation of SQL optimizer, free of complex analysis tuning the statement.
- PB-level data analysis in seconds.
MPP horizontal scaling architecture to support PB-level data query-second response. Calculated to quantify intelligence and column store index, the leading traditional database engine performance 10x.
- Availability, service is always online.
Support distributed transactions, data consistency ACID support, across all nodes and data redundancy machine deployment, automated monitoring switch, always-on service.
- Wide ecological compatibility.
Support for mainstream BI, ETL tools. GIS data analysis through PostGIS plug-in support, MADlib library built more than 300 machine-learning algorithm library.
- Data interoperability.
Supported by DTS, Dataworks tools such as synchronization with multiple data sources; support high concurrent access to OSS, data analysis and build the lake.
Product Architecture
AnalyticDB for PostgreSQL using MPP architecture, instances of the plurality of computing the group consisting of, linearly expandable. Includes components are:
- The coordinating node (Master Node).
- Receiving a request to develop a distributed implementation plan
- Computing Group (Compute Groups).
- Full analysis of parallel computing
- Double copy of the data storage
- Regular automatic backups to OSS