Table of contents for Teradata Influencers Summit 2015
- Teradata Technology Innovation #TD3PI
- Teradata Marketing and MGM #td3pi
- Teradata Strategy Overview #TD3PI
- Teradata Unifed Data Architecture #TD3PI
- Teradata Listener #TD3PI
- Think Big Consulting Update #TD3PI
- Think Big Data Lake Program #td3pi
- Real Time Big Data #TD3PI
- Teradata Aster Comments #TD3PI
- Teradata Aster AppCenter #td3pi
Randy Lea came back on o talk about Teradata Aster. He started by re-emphasizing what he sees as the value of big data and big data management. Teradata, he says, has a strong infrastructure for acquisition, integration and access to data. Adding Big Data into this environment is about allowing data to be landed quickly in the data infrastructure with only light transformation and integration while also allowing rapid experimentation and access to this data. The overall environment should allow full modeling, partial modeling or modeling on read as needed. In addition most customers spend too much of their capacity on ETL processes. This can be cleaned up and then you can add a well managed data lake to handle the rest of the data, the data that wasn’t making it into the data warehouse.
The core value add, though, is about driving analytic value from this data not just storing the data and managing it more effectively. This means making more sophisticated analytics available against this data but historically lots of fragmented tools made this hard. What you need, he says, is an integrated interactive analytic environment that supports statistics, graph, machine learning, etc.
He re-iterated that you need to begin at the business process – you have to drive changed decision-making into an operational process to add value with analytics. In general this means making predictions about customers or other entities using statistics, machine learning, text or behavioral time series data. There’s no one best answer, you need support for all of them. These different types of “big analytics” need to be applied to a wide range of data coming from a wide range of systems and processes.
Teradata Aster is designed to provide these different analytical engines in an integrated way and Randy gave some interesting examples of using test and path analytics together for instance. The SNAP framework for Teradata Aster combines the various engines (SQL, Graph etc) so they can be used in a single operation in an integrated way. 120 pre-built analytical functions complement these engines to make it as easy as possible to develop these analytics.
In 2015 Aster focused on doubling their sales organization, offering new solutions (analytic solutions, analytics as a service), more assets for self-training on the Aster Community and AppCenter. AppCenter allows potentially complex analytics to be parameterized and provided as pre-packaged apps for Teradata Aster users. These can be configured and executed by users without writing any SQL MR code.
Fundamentally Randy sees a shift from requests for a big data strategy to requests for an analytic strategy.