Miguel Liroz-Gistau has received the best presentation award from the Grid5000 Spring School 2014 in Lyon for his talk on “Using Grid5000 for MapReduce Experiments” (Miguel Liroz-Gistau, Reza Akbarinia, and Patrick Valduriez).
Abstract of the talk:
MapReduce is one of the most popular solutions for big data processing. In our recent research activities, we have improved the MapReduce framework by enhancing data locality and load balancing during the MapReduce job executions. Particularly, we developed two prototypes: 1) MRPart for reducing the data transfer between map and reduce nodes; 2) FP-Hadoop for bringing more parallelism to the framework and balancing the load of reduce nodes. We used Grid5000 for evaluating the performance of our solutions. In this paper, we describe our methodology for deploying and testing the developed prototypes in Grid5000.