Skip to content

Latest commit

 

History

History

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
 
 
 
 
 
 
 
 

README.md

Performance test for Block Slicing

Usage

  1. $ pip3 install ./program_slicing
  2. $ cd scripts/performance_benchmarks
  3. $ pip3 install -r requirements.txt
  4. $ cd block_slices
  5. Script parameters:

--dir folder with java files

-i number of iterations to run all files, default is 1

Run $ python3 check_performance.py --dir dataset -i 2 where dataset is teh folder with java files and number of iterations is 2.

You should have access to global network to use pip. Python 3.8 with corresponding C compiler is required. Run Python Console to check the version of C compiler.


Last reported result

PC info:

RAM 64GiB

CPU Intel(R) Xeon(R) Gold 6266C CPU @ 3.00GHz:

The algorithm was run for 15 java files with 100 ncss. The procedure was run 1 time(s) for more accurate calculations.

Total time of running 15 java methods is 32.0 secs.

Average time for 1 method: 1.667 secs.

Min time of 1 method: 0.000 secs,

max time of 1 method: 11.000 secs,

median: 0.000 secs,

quantile 75%: 1.500 secs,

quantile 95%: 7.500 secs