A bottleneck limits a computation’s throughput. In a coin-counting task, one worker becomes the bottleneck when it must sum hundreds of long lists. section 13.2.1
Adding an intermediate aggregation level divides the results among workers, each summing shorter lists. This redistributes the work; the example does not diagnose a memory-specific limit. section 13.2.1
Figure 13.5 Revised coin counting process, using MapReduce and a little help. Now, each counting machine produces several lists, one for each member of the McDuck family at level 2. They, in turn, will have to check a hundred lists, but each with just a few entries, and sum up the values in each list by entry.
The revised coin-counting diagram shows how an intermediate level distributes summation work. section 13.2.1
