JethroData indexes every column. Queries read only the rows they need instead of performning a full-scan.
JethroData's performance directly relates to the size of the result-set, not the size of the dataset. The deeper the user drills-down, the faster it gets.
As new data continuously streams in, Jethro makes it immediately available for queries. Users can view live dashboards and detect trends as they occur.
Jethro supports standard SQL and easily integrates with BI tools using JDBC/ODBC. Jethro’s support for complex SQL JOINs is unparalleled in its ability to perform JOINs on large tables.
JethroData works with any standard Hadoop distribution and can be easily added to an existing Cluster. It runs on separate compute nodes and no JethroData components are installed on Hadoop nodes.
With the unlimited storage scale-out of HDFS, and JethroData's elastic query nodes, JethroData can support the ingestion of billions of events per hour, fast queries of TB-size tables, and hundreds of concurrent users.