Terracotta recently announced the BigMemory-Hadoop Connector product which allows Hadoop to write directly to memory to execute jobs quicker and provide results faster.  The BigMemory-Hadoop Connector will by pass the file system (HDFS) and will instead write to memory which is much faster.  Clients can also read the results from BigMemory using BigMemory-Hadoop Connector which will bring Hadoop job results to the customer faster than before.

Hadoop isn't a silver bullet, however, It's innovations like these that will further propel the world distributed computing and data analytics forward into the enterprise and beyond.

You can read about the BigMemory-Hadoop Connector as well as download the project here. 

Ford has been around a long time and they've always been known as an innovators.  Much to my surprise, Ford's CEO Alan Mulally began a campaign in 2006 to use machine learning and data analytics throughout their entire organization to turn the company around and sustain their new found success.  "Analytics and operations research was a major enabler of our turnaround and our ongoing success as a data-driven company," said  Bob Shanks, Ford Motor Company executive vice president and chief financial officer.  "We continue to build on our history of data-driven decision making," said John Ginder, manager, systems analytics and environmental sciences in Ford's Research and Innovation group. "Throughout the company, we're using data science to create smarter business strategies, make better product decisions, assist dealers to be more successful, and improve customer satisfaction.".  Ford has used predictive models to determine the products that they will build.  "Social media and the vast amount of online conversation is helping us get a faster and more specific data set to help us make product decisions. We now use text-mining algorithms to formulate a more complete picture of what consumers want that is not available using traditional market research," said Michael Cavaretta, Ford technical leader for predictive analytics and data mining.

Read more here http://www.heraldonline.com/2013/04/08/4755620/ford-wins-2013-informs-prize-for.html

Mobile advertising based on machine learning

ADTHEORENT's Real-Time Learning Machine (RTLM) is a break through in machine learning predictive modeling.  The product analyzes information related to an advertiser and creates a prediction model to suggest profitable ads in real time!  The white paper claims that the technology is able to process huge data sets in a fraction of the time that it takes for other machine learning tools. 

You can read all about this new machine learning mobile advertising tool in the official white paper here.

I can't comment on the efficiency of the product until I've tried it, but if it does what it's advertised to do, then this is a must have for mobile app developers and advertisers.

What do you think?

What is a bytepusher?

byte·push·er

  [bahyt-poosh-er]