Balancing reducer workload for skewed data using sampling-based partitioning

2014 
Display Omitted MapReduce-based systems process skewed data inefficient and incur load imbalance of all reducers.We design a sampling method to process the dataset and give a theoretical guarantee ...
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    0
    Citations
    NaN
    KQI
    []