README

########## SETUP ##############
To setup for running the benchmarks you must first create simlinks to the following node executables in the nodes directory

nodes/node_watchdog/node - the executable that contains the modifications to node to implement the watchdog functionality.
	The version with the watchdog changes can be created by cloning and checking out HEAD.
		`$(git clone ../.. nodes/watchdog_node && cd nodes/watchdog_node  && ./configure && make -j $(nproc) && cd -)`


nodes/original_node/node - the executable at the node commit dc6bbb44da before any node-cure commits were made.
	The version with no watchdog changes can be created by cloning and checkign out commit dc6bbb44da.
		`$(git clone ../.. nodes/original_node && cd nodes/original_node && git checkout "dc6bbb4" && ./configure && make -j $(nproc) && cd -)`

Once the executables have been created/linked the benchmarks are ready to be run.



####### Running Benchmarks ######

#NOTE when running benchmarks the environment variable NODECURE_SILENT should be set to remove unnessesary overhead due to logging.
Aslo ensure if you run with sudo that environment variables are preserved.

The benchmarks can be run with the script `run_benchmarks.sh`.
Once the benchmarks have completed running, the results will be stored in the result/<test_name>.result and the results_single_cpu/<test_name>.result.



This script will generate a new random file for use in the randomRead.js test then run every tests against both the watchdog_node and the original_node.

#TODO make the benchamrks run on every node executable in nodes/*/node so that way any additional changes can be tested without changing th ebenchmark scripts.