Provided by: Association for Computing Machinery
Date Added: Aug 2013
Iterative computations are widely used in cloud intelligence applications, such as the well-known PageRank algorithm in web search engines, gradient descent algorithm for optimization, and many other iterative algorithms for applications including recommender systems and link prediction. Cloud intelligence applications often perform iterative computations (e.g., PageRank) on constantly changing data sets (e.g., web graph). While previous studies extend MapReduce for efficient iterative computations, it is too expensive to perform an entirely new large-scale MapReduce iterative job to timely accommodate new changes to the underlying data sets.