BI queries on Hadoop
become lightning fast!
Query Big Data with immediate responses
is a BI analytical engine that responds in milliseconds regardless of data volume.
These extreme performances are achieved using multidimensional indexes and pre-calculated aggregates updated in real time.
Connect your data visualisation tool directly on Hadoop
Instantly navigate data from a Hadoop cluster using data visualization tools such as Tableau®, QlikView® or MicroStrategy.
With just one click, your big data becomes easily available to all users without having to set up complicated ETL processes to aggregate data.
Scalable to petabytes and thousands of users
Indexima can index petabytes of data with a smaller number of servers. So, very high volumes of data can be processed at a lower cost.
During performance testing, Indexima achieved over 3,000 queries per second. When used by our clients, the data visualisation tools normally send around thirty simultaneous queries for each connected user with no perceptible loss of performance.
Insert and delete data easily
Indexes and aggregates automatically update after every insertion or deletion.
You can quickly index large volumes of data on an extensive scale or index continuously with real time insertions.
You can easily delete data by specifying any criteria.
Deploy Indexima with or without Hadoop
Indexima can be deployed on a cluster in two ways:
  • As a Yarn application, Indexima is automatically deployed by Hadoop on all the cluster nodes.
  • Indexima can also operate independently as a "standalone", if you are not using Hadoop.
Advanced SQL support
Groups, joins, subqueries: Indexima supports queries generated by the market's main data visualisation tools.
How does it work?
Indexima offers a new approach to querying big data by converging several technologies (column-oriented distributed engine, OLAP cubes, multi-column indexes, in-memory, materialised views) into a single tool.
The multi-column indexes are distributed and persisted in "in-memory" when RAM permits. Like an OLAP cube, they include pre-aggregations. This lets it respond to most queries in milliseconds, regardless of the data volume.
When the query relates to certain unindexed columns, the column-oriented engine accesses data on the disk by only loading the necessary blocks by means of other indexed columns.
Count distinct aggregates may be included in the indexes and these guarantee very fast responses by returning an exact value without approximation.
Indexima can thereby respond effectively to any type of query:
global aggregation queries concerning all the data or queries with a precise where clause concerning a data subset.
Try Indexima
Free trial
Try now  ►