“And now you know the rest of the story.” – Paul Harvey
Paul Harvey was an American radio broadcaster who ended each of his ABC radio broadcasts with the tag line, "And now you know the rest of the story." His idiosyncratic stories always seemed to dig up the details that nobody had heard.
In a similar fashion, the overwhelming growth of big data for enterprises today means that not only must you be able to analyze the rest of the information that truly counts, but that involves sophisticated search and machine learning algorithms over large data sets to find the important details that matter.
Oracle is re-thinking this challenge with their innovative approach to processor design.
In March 2016 Oracle announced a free and open API and developer kit for its Data Analytics Accelerator (DAX) in SPARC processors through its Software in Silicon Developer Program. The SPARC M7 DAX is a unique innovation that accelerates a broad base of industry-leading analytic applications to help solve big data challenges.
DAX accelerators were specifically designed to accelerate analytical queries, and were initially supported in the Oracle Database 12c In-Memory option. This open API for DAX is designed to expand the existing program so application developers can leverage the DAX technology to accelerate a broad spectrum of software applications, including big data analytics, machine learning, and more. Apache Spark's in-memory framework is an ideal showcase for demonstrating this kind of acceleration benefit of DAX (more detail: Apache Spark and DAX).
Enter stream processing, which can be a game changer in a big data world.
This technical article describes stream processing using the DAX APIs in detail:
Introduction to Stream Processing Using the DAX APIs
These DAX APIs allow you to use stream-processing techniques to analyze and act on real-time streaming data that is in-memory, by taking advantage of the DAX hardware acceleration on the SPARC microprocessor. Stream-processing techniques allow efficient use of system resources by structuring memory operations as regular patterns that can be accelerated by the DAX co-processors.
The DAX co-processors enable direct execution of many of the DAX API operations, manipulating in-memory data streams directly and freeing the SPARC CPU for other tasks. The DAX APIs will continue to evolve over time as more applications are developed and more analytic functions are added.
Developers interested in accelerating analytics with the open API for DAX can register to access to Oracle’s Software in Silicon Cloud at: http://swisdev.oracle.com/.
This 5-minute video demonstrates how DAX work: Oracle’s Data Analytics Accelerators (DAX)
For more information, or to enter a discussion on this topic,
visit our Software in Silicon Community page on the Oracle Technology Network:
Develop the Next Generation of Analytics and Security
And now you do know the rest of the story!