Aster Data Delivers 30 Analytic Packages and MapReduce Functions for Mainstream Data Analytics

Updated: June 21, 2010

Traditional data management platforms and analytic solutions do not scale to big data volumes and restrict business insight to views that only represent a sample of data, which can lead to undiscovered patterns, restricted analysis and missed critical events. MapReduce is emerging as a parallel data processing standard, but often requires extensive learning time and specialized programming skills.

Coupling the SQL language with MapReduce eliminates the need to learn MapReduce programming or parallel programming concepts. Other benefits of this coupling include:

  • Making MapReduce applications usable by anyone with a SQL skill-set.
  • Enabling rich analytic applications to be built in days due to the simplicity of SQL-MapReduce and Aster Data's suite of pre-built analytic functions.
  • Delivering ultra-high performance on big data, achieved by embedding 100 percent of the analytics processing in-database, eliminating data movement.
  • Automatically parallelizing both the data and application processing with SQL-MapReduce for extremely high performance on large data sets.

New functions

Aster Data also announced today a significant expansion in the library of MapReduce-ready functions available in Aster Data nCluster. The Aster Data nPath function is only one example of more than 1,000 functions now delivered through over 40 packages available with the Aster Data Analytic Foundation for Aster Data nCluster 4.5 and above.

These new functions cover a wide range of advanced analytic use cases from graph analysis to statistical analysis to predictive analytics, that bring extremely high value business functions out of the box that accelerate application development. Examples include:

  • Text Analysis: Allows customers to "tokenize" count and position or count the occurrences of words as well as track the positions of words/multi-word phrases.
  • Cluster Analysis: Includes segmentation techniques, like k-Means, which groups data into naturally occurring clusters.
  • Utilities: Includes high value data transformation computations. For example developers can now simply unpack and pack nested data as well as anti-select, or allow the return of all columns except for those that are specified.
Featured Research