CPU_Aggregation

Data Generation

• With root privileges you should run datasetgen.sh which is under datagen folder, it takes two arguments $datasize in GB(1, 10 or 100) and $skew of data (0 for uniform 1.05 for low skewed and 1.25 for high skewed data). This script will create data files under /slowdisk folder.

    ./datasetgen.sh 10 0

Running the aggregation

• You should run sudo apt-get install libbz2-dev first.
• Then run run.sh and give the data size in GBs as parameter.

    ./run.sh 10

• The results file will be created for each different scheme under the folders standalone_aggreagate_{scheme}_{hash_function}