MAP REDUCE
Map Reducing is a two-step process in which massive amounts of unstructured data are placed in clusters to be analyzed. The first part of the process is mapping in which the data is filtered and sorted. The second step involves a summary operation of the filtered data which is the reducing part of the process. Map Reducing is important for the data analytic industry because it lays the framework for how to get usable information out of unstructured data.
During the semester in our Data Analytics class, we encountered many situations in which the data was in front of us but we had no idea where to filter and find the data we needed. In the technology era, there is an insane amount of data being misused or not even being used at all. Map Reducing culminates all the topics we have talked about because each topic relates back to filtering and sorting information. For example, Map Reducing builds upon SAS clustering because they both include the process of reducing the data into clusters to make data more usable.
Map Reducing is changing the way companies filter data because it lays the framework for grouping the data into separate processes. This function is extremely useful for companies dealing with large amounts of data because it divides the data into many smaller parallel processes. For example, Teradata.Aster uses SQL-MapReduce to let companies perform many different processes they were unable to do before MapReduce. Aster data conducted different case studies using this technology but the one that stuck out to me was about fraud detection in online gaming. Before implementing SQL-MapReducing, fraud detection was once a week because the data was too large, also, the fraud was detected too late or never at all. After implementing this technology, fraud detection improved in various ways. First, the overall time of detection was reduced from 1 week to 15 minutes. Secondly, the fraud detection system was catching fraud that previously was not being recognized. Furthermore, Map Reducing will enhance the way data interpretation is done.
Works Cited
http://www-01.ibm.com/software/data/infosphere/hadoop/mapreduce/
http://searchcloudcomputing.techtarget.com/definition/MapReduce
http://www.asterdata.com/resources/sql-mapreduce-casestudy-fraud/index.php