When: Wednesday, 2015-May-06, 11h30-12h00
Where: FCUL-DI, room C6.3.38
Presenter: Pedro Costa
Abstract:
MapReduce is often used for critical data processing, e.g., in the context of scientific or financial simulation. However, there is evidence in the literature that there are arbitrary (or Byzantine) faults that may corrupt the results of MapReduce without being detected.
In this smalltalk I will present an (i) overview of the MapReduce framework, (ii) focus on the limitations of the framework, (iii) describe the existent related work, (iv) show a new design to make the framework able to tolerate arbitrary faults, (iv) present two modes of execution in order to improve performance, (v) and show the results of the experiments.
This is a first step in order to scale out MapReduce computation to a cloud of clouds environment.
About the presenter: PhD student