Why Trifacta

Interactive Exploration

Trifacta presents the user with automated visual representations of their data based upon its content. These profiles automatically present elements of the data in the most compelling visual representation – geographic elements are presented as maps, time-oriented elements are presented according the common hierarchies such as day, month, year, etc.

Every profile is completely interactive – allowing the user to simply select certain elements of the profile to prompt transformation suggestions.

Predictive Transformation

Trifacta’s visual representations of data are interactive – enabling the user to click, drag or select over the specific attributes of the data they’d like to manipulate. Every interaction within Trifacta leads to a prediction – the system evaluates the data you’re working with and the interaction applied to recommend a ranked list of suggested transformations for the user to evaluate or even edit depending upon what they’re trying to do.

As users browse through the different suggested transformations presented to them, the system presents a preview of how each transformation will impact the data itself. This iterative feedback loop is always occurring throughout the use of Trifacta- constantly taking inputs from the data and the user to intelligently recommend ways to manipulate the data.

Intelligent Execution

Every transformation step defined by the user in Trifacta is logged and at execution time automatically compiles down into the appropriate processing framework based upon the scale of the data the user is working with and the type of transformations being applied.

Depending upon these inputs, Trifacta can compile down to MapReduce, Spark and our own single node execution engine for smaller data sets. This is all done behind the scene – abstracting the user from the underlying execution framework.

Collaborative Data Governance

To meet the growing data governance requirements of modern IT departments, Trifacta provides support for collaborative security, access, data lineage and metadata.

• Security – rather than implement a separate security framework, Trifacta builds upon existing Hadoop security standards, including Kerberos. This allows IT departments to manage security within Trifacta and Hadoop in parallel, eliminating the need for a separate framework.
• Metadata & Lineage – through integrations with Cloudera Navigator, Apache Atlas, Hive metastore and HCatalog, Trifacta can publish and provide visibility to metadata and the lineage of data created from the wrangling process.
• Operationalization – Trifacta supports enterprise schedulers Chronos and Tidal enabling operational transformations created in Trifacta to run on specific schedules, according to the production requirements of organizations.