This code is fairly new. Please contact me (gkt@cs.luc.edu) if you have any questions.
You will need Apache Spark to run this code.
usage: spark_benchmark_hpc.py [-h] [-g] [-b BLOCKS] [-s BLOCK_SIZE] -n NODES
[-p NPARTS] [-c CORES] [-j JSON]
Simple Map Spark Microbenchmark - Python Version
optional arguments:
-h, --help show this help message and exit
-g, --generate generate data (boolean)
-b BLOCKS, --blocks BLOCKS
number of blocks
-s BLOCK_SIZE, --block_size BLOCK_SIZE
block size
-n NODES, --nodes NODES
number of nodes (for reporting)
-p NPARTS, --nparts NPARTS
how many partitions to create per node
-c CORES, --cores CORES
number of cores per CPU
-j JSON, --json JSON where to write the report