I am in a situation where I am using Apache Spark and its map/reduce functionality. I am now at a stage where I have been able to map to a data set that conceptually has many "rows" of data.
Now what I am needing is to do a reduce which usually is a straight forward thing. My real need though is to reduce on "overlapping" rows. For example, the first reduce uses "rows" 1-30, the second uses 11-40, the third 21-50 and so on. How would this work in a Spark environment? I appreciate any insight or directions anyone can give, Jeff Richley