Difference between revisions of "CSC352 MapReduce/Hadoop Class Notes"

From dftwiki3
Jump to: navigation, search
Line 137: Line 137:
 
[[Image:MapReduceDataFlowOneReduce.png]]
 
[[Image:MapReduceDataFlowOneReduce.png]]
 
</center>
 
</center>
 
+
The shaded boxes are nodes.  The dotted arrows show transfers on a node.  The heavy arrows show transfers across nodes.
  
 
<center>
 
<center>
 
[[Image:MapReduceDataFlowTwoReduces.png]]
 
[[Image:MapReduceDataFlowTwoReduces.png]]
 
</center>
 
</center>
 
+
The general case of having several reduce tasks.  In this case the outputs of the map tasks are shuffled; each reduce task receives many outputs of map tasks.
  
  
Line 148: Line 148:
 
[[Image:MapReduceDataFlowNoReduce.png]]
 
[[Image:MapReduceDataFlowNoReduce.png]]
 
</center>
 
</center>
 
+
It is possible to have 0 reduce tasks...
  
  

Revision as of 18:18, 31 March 2010


This section is only visible to computers located at Smith College