Enabling fast failure recovery in shared Hadoop clusters: Towards failure-aware scheduling | doi.page