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