Analytical Databases edit  

Our list of and information on commercial, open source and cloud based analytical databases, including Teradata, Exadata, Redhift and alternatives to these.

Category Definition

Full stack databases (supporting both storage and query of data) that focus on analytical or OLAP use cases that generally involve large scanning or aggregation operations. Typically support distributed parallel execution of queries (and are therefore commonly referred to as MPP databases) with columnar compression, and often support a range of analytics beyond SQL queries, for example cube based MDX queries, machine learning, graph or geographical analytics. Some technologies also support a level of query federation using external tables (for example over data in Hadoop). Some technologies run over Hadoop (exploiting HDFS and YARN), but will either use their own proprietary data format or will be positioned as a self contained analytical database.

Further Information

The following analyst material covers a number of technologies in this category::

See our Query Engines page for details of technologies that support similar capabilities but over external data (e.g. data in HDFS or source databases).

Commercial Technologies

IBM Db2 Warehouse (formerly dashDB for Analytics)IBM Db2 and BLU (in memory) for Docker container supported infrastructure (also available as an appliance and cloud service - see below) -
Teradata DatabaseMPP database with support for a range of data warehouse and analytics functions; deployable on private or public cloud (also available as an appliance and a cloud service - see below) -
Teradata Analytics PlatformAnalytics platform that supports graph, text and IoT analysis plus machine learning (also available as an appliance and a cloud service - see below) -
VerticaMPP columnar database with support for a range of analytical functions including machine learning; deployable on commodity infrastructure or public/private cloud -
Actian VectorMPP columnar database with support for vectorized execution, small incremental inserts and min-max indices, with a free community edition for databases under 1 Tb -
InfoBrightDBColumnar database, now sold by Ignite Technologies -
SQreamColumnar GPU accelerated analytical database, available on premise or in the cloud -

Open Source Technologies

GreenplumMPP database based on PostgreSQL, with support for multiple storage models and analytical capabilities (including graph); open sourced in October 2015
MariaDB ColumnStoreColumnar storage for MariaDB (the open source fork of MySQL) based on a fork of InfiniDB -
MariaDB AXData warehousing solution based on MariaDB ColumnStore, with commercial support available from MariaDB -
MonetDBOpen source columnar database - ;
InifiDBOpen source columnar database, inactive since March 2015 -
AresDBGPU powered real-time analytics db from Uber -
Apache Pinot (incubating)Open source realtime distributed OLAP datastore from LinkedIn -;

Hadoop Based Open Source Technologies

Apache HiveAn analytical database when used with LLAP, ORCFile and Tez; runs over Hadoop
Apache ImpalaAn analytical database when used with Kudu or Parquet over HDFS; runs over Hadoop
Apache HAWQA port of the Greenplum MPP database (which itself is based on PostgreSQL) to run over Hadoop
Apache TajoDistributed analytical database engine that runs over Hadoop
PrestoAn analytical database when used over Hive/Hadoop , originally created and open sourced by Facebook -
DruidOLAP database supporting real time aggregations of streaming data using HDFS/S3 as backing storage

Hadoop Based Commercial Technologies

Vertica on HadoopVertica running on Hadoop -
Actian Vector HVersion of Action Vector that runs as a native YARN app but requires data to be loaded into its proprietary data format -


Oracle ExadataAn appliance consisting of an Oracle RAC cluster combined with a set of storage nodes via high bandwidth interconnect, with support for hybrid columnar compression -
Microsoft Analytics Platform SystemAn appliance built around SQL Server Parallel Data Warehouse and PolyBase -
IBM Integrated Analytics SystemAppliance built around Db2 Warehouse and BLU (in memory) acceleration with support for Spark -
IBM PureData System for Analytics (formally Netezza), now discontinuedAppliance utilising FPGA chips to run elements of queries in hardware, with support for a range of languages including R -
Teradata IntelliFlexTeradata capabilities as an appliance -
Teradata IntelliBaseA mixture of Teradata and Hadoop nodes in a single appliance -
Pivotal EMC Data Computing Appliance (DCA)Greenplum appliance -

Cloud Services

Amazon RedshiftA MPP analytical database, with support for columnar storage and the ability to query data in Amazon S3 as external tables (Redshift Spectrum) -
Google BigQueryAnalytical SQL database service, with cost based on storage and query execution -
Azure SQL Data WarehouseScalable analytical database, with support for Azure Data Lake Store external tables -
Cloudera Altus Data Warehouse - Apache Impala as a cloud managed service over AWS or Azure 
Oracle Exadata CloudOracle Exadata as a managed cloud service (including as an Oracle managed on premises offering) -
IBM Db2 Warehouse (formerly dashDB for Analytics)IBM Db2 and BLU (in memory) acceleration as a cloud service -
Teradata IntilliCloudTeradata Database, Hadoop and Aster as a service -
SnowflakeData Warehouse for the cloud, with separated compute and storage, columnar storage, vectorized execution, adaptive optimisation (no indexes, keys or tuning required) and support for semi-structured (JSON, Avro and XML) data -

Analytical In Memory Technologies

MemSQLDistributed in memory relational database, with wire compatibility with MySQL and support for row and columnar storage, and a free community edition -
SAP HANAIn memory relational DBMS primarily focused on accelerating SAP applications -
EXASOLIn memory MPP database with columnar compression and SQL support -
MapDIn memory, column store, SQL relational database that runs on GPUs -
KineticaDistributed in memory relational database that runs on GPUs -
CubeDBA open source simple but fast in-memory multi-key counter store from Badoo -

Analytical Graph Databases

DataStax EnterpriseCommercial product built on Apache Cassandra with the addition of graph and search capabilities -
TigerGraphCommercial hybrid OLTP/OLAP graph database that claims order of magnitude performance and scalability improvements over it’s competitors; previously known as GraphSQL -,
AnzoGraphMassively parallel distributed graph database built for analytics -
GraphBaseCommercial graph database designed for use in AI applications -
JanusGraphOpen source distributed graph database that runs over a number of storage backends (including Cassandra, HBase and BigTable), with TinkerPop support including support for graph analytics; previously known as Titan -
TinkerGraphIn memory graph databases that’s part of TinkerPop as a reference implementation -
GRAKN.AIOpen Source graph database designed for AI use cases that also supports graph analytics -
Complexible StardogRDF database that also support property graphs and data virtualisation, with a community edition available -
Trovares xGTMassively parallel analytical in memory graph database from Cray alumni -

Blog Posts