What's new in Aster Analytics?

Learn Data Science
N/A

What’s New in Aster Analytics Portfolio 7.0?

Aster Analytics Portfolio 7.0 includes more than new functions, but also offers new interfaces for the analytic community.  This is a significant release allowing a broad analytic community to use Aster Analytics to uncover insights to impact business outcomes.

Let’s take a step back to level set on the various components of Aster Analytics. Starting at the bottom of figure 1 is the brawn of Aster Analytics - the Aster database and Aster Execution engine that provides the underlying MPP architecture.  The Aster database runs on commodity hardware, Aster Appliance and both public and Teradata Managed Cloud. The Aster Execution Engine supports the Hadoop implementation.

 

Next layer is the brains of Aster Analytics – Aster Analytics Portfolio that includes 100+ prebuilt analytic functions for data access, data preparation, analytic modeling and visualization. The portfolio also includes interfaces to the KNIME workbench and Aster R. 

The top level represents the tools and apps - the beauty of the Aster Analytics. These include AppCenter Apps and other user interfaces such as workbenches, BI and visualization tools. 

This blog focuses on the Aster Analytics Portfolio – the brains. What’s new?

  • Workbench for Aster Analytics: KNIME Analytic Platform now supports Aster Analytics functions. As part of Aster Analytics portfolio 7.0, Teradata developed the Teradata Aster Extension for KNIME – software that adds Aster functions into the KNIME interface. KNIME provide a drag and drop workflow editor to simplify and accelerate model development. Data scientist, business analyst and citizen data scientist can all leverage the KNIME interface to combine KNIME and Aster Analytics functions to build a complete workflow from data access and preparation to model development and visualization. And you don’t have to be a SQL expert to use Aster Analytics.    

KNIME is open source software and can be downloaded from https://www.knime.org/downloads. Aster Extension for KNIME downloads, videos and blogs are coming to the community soon. If you don’t want to wait, you can contact your Account Executive for access to the Teradata Aster Extension for KNIME.

  • Aster R:   All Aster Analytics functions are now available through R. This extends the use of Aster Analytics to R programmers, one of the fastest growing analytic communities. R programmers have access to 164 Aster Analytic functions and over 100 SQL wrappers for additional R functions and operators. Analyst can continue to use their favorite R client interface to access Aster’s multi-genre advanced analytics at scale. And don’t forget, you can also embed the R engine within the Aster MPP environment to easily scale “embarrassingly parallel” tasks such as scoring and other row independent tasks.
  • Aster to Spark Connector: Aster Analysts can easily connect and incorporate Spark analytics into the Aster workflows. This connector and wrappers simplifies the use of Spark and operationalizes Spark analytics through Aster AppCenter. Aster Spark Connector blog describes how to use the Aster Spark Connector, “RunOnSpark” function to execute the Spark function LinearRegrWithSGD function.
  • New analytic functions: Teradata added new analytic methods that support IoT use cases and model validation. IoT functions include the Fast Fourier Transforms that is used to analyze sensor patterns to find anomalies and similarities. Model Validation function: ROC calculates values for a range of discrimination thresholds: True Positive Rate (TPR), False Positive Rate (FPR), Area under the ROC curve (AUC) and Gini Coefficient, allows the analyst to measure the accuracy of their model.

Aster Analytics portfolio 7.0 is compatible with Aster Analytics 6.20 and 7.0, across all deployment options.

Aster Analytics Portfolio (A.K.A. Foundation) 7.0 is available today. Talk to your Account Executive to upgrade today.