Product Review: Teradata Aster Analytics
Teradata Aster Analytics is a multi-use analytics platform designed for businesses desiring an integrated solution for their analytics needs. The platform provides end-to-end support for all things related to big data, including warehousing and data visualization. The Aster platform also supports multiple analytics techniques, analytics frameworks and deployment options. Since 2011, when Teradata acquired Aster Data Systems for $263 million, Teradata has continued to expand the capabilities of its major enterprise analytics platform.
One major selling point for the Aster is its ability to use a single SQL query to generate multi-genre analytics and help businesses gain new insight. This is helped by Aster’s use of a single framework using Graph, MapReduce and R. This, combined with massive parallel processing (MPP) allows the Aster to analyze data very efficiently, even for large data sets.
The Aster Analytics platform gives business executives the tools to perform a number of functions for data organization and analysis. On a more specific level, for analysts and data scientists, the Aster platform is designed to support five different methods for pursuing data analysis. These are path analytics, text analytics, connection analytics (via a graph processing engine), machine learning, and IoT analytics. Within these five categories, there are several individual items of interest for enterprises.
Related: Virtualization Manager, helps administrators manage resources, capacity and performance in the virtual environment
One of these is Teradata Aster Customer Satisfaction Index, which is powered by the Aster Analytics platform. The Customer Satisfaction Index is enabled by the text analytics capabilities of Aster and is designed to help companies understand customer behavior. In addition, the CSI allows companies to study customer decision-making with some sense of cause-and-effect, or at least an attempt to find which events lead to certain customer outcomes. The CSI makes use of Graph and Customer Path Analytics, also part of the Aster platform, as well as the text analytics mentioned previously. Results are analyzed via the Aster AppCenter, a web-based application for data visualization, or through a third-party visualization tool.
Another key feature of the Aster platform is its graph analytics, also called Teradata’s Connection Analytics. This class of analytics looks for relationships between people, processes and products – in other words, it is aimed at bringing data from multiple disciplines together and looking for patterns. Within Teradata’s Connection Analytics, the Aster SQL-GR is an important component. This tool is a graph analytics engine that includes graph processing and large-scale analysis capabilities. The SQL-GR uses SQL to handle the data, but does not presuppose that an analyst will have a specialized knowledge of the language; Teradata advertises that a single query may be all that is needed to perform graph analysis.
Within the last few years, Teradata has also added support for the R programming language via the Aster R feature. While R has become a very popular choice for data mining in the last several years, standard tools for the language are not always scalable to large data sets. Using R can be resource-intensive, which is why the Aster R tool uses Teradata’s massive parallel processing (MPP) architecture to maximize performance. Aster R makes use of preconstructed R functions in the Aster Portfolio, but also harnesses the power of open-source R libraries for expanded functionality.
Related: Network Configuration Manager, bulk management and reliable backups for network device configurations
While the Teradata Aster platform has a number of solid features, no product review would be complete without a look at a few of its competitors. The Apache Hadoop framework has been named as a competitor, though this is not a standard competition scenario; a 2012 Teradata Magazine article entitled “Head to Head Competition” notes in the subtitle that “Teradata Aster and Apache Hadoop play complementary roles in Big Data analytics and processing.” While both platforms make use of MapReduce, Hadoop uses its own file system (the HDFS) and Teradata makes use of a massively-parallel processed (MPP) relational database. The article goes on to point out that both products have their own strengths and weaknesses, and that in some cases, both Hadoop and Aster could work together, particularly through Teradata Aster’s SQL-H tool. A few years later, in 2015, the Aster Analytics on Hadoop solution was made available, allowing for further collaboration between the two systems.
More generally, Teradata competes with a number of large companies in the data management market, including Oracle, Microsoft and IBM. A number of these companies were mentioned in Gartner’s 2016 Magic Quadrant report for Data Management Solutions and Analytics, which is made available to the public through Teradata’s website.
Teradata Aster Product Page
Teradata Aster Community
Teradata’s Upcoming Events
Teradata Asset Library