stressypy 0.0.9

A simple program for calling stress and/or stress-ng from python

stressypy uses the unix package stress to stress a certain number of cpus for a certain amount of time, as specified by the user. It creates JobBlock objects which contain pertinent information for using these stress loads to test queueing algorithms.

JobBlock Attributes:

class attributes

• total_blocks: the number of blocks created

• waiting_blocks: the number of blocks waiting to be queued - in box

• queued_blocks: the number of blocks waiting to be executed - in queue

instance attributes

• n_cpu: number of cpus being stressed

• t_run: the time it will take to run the job

• delta_t_run: the time the processor will take to run the job; overhead time

• state: the state of the block

• func: the function the block is storing

• func_args: the arguments for the function the block is storing

• area: the ‘area’ as width in cpu vs height in time

• queue: the queue the block will be sent to

• job: a combination of the func and arg to return the complete job that the block should execute

Installation

stressypy can be installed with pip install stressypy

or cloned manually and setup with python setup.py install

stressypy is dependent on the stress unix package. Make sure you have it installed.

Unix Distro	Command
Debian	sudo apt-get install stress
Arch Linux	pacman -S stress

Directions

stressypy runs using the command stressy stress with the number of cpus and time passed as arguments

• stressy stress 1 1 stresses 1 core for 1 second

• stressy stress 7 3 stresses 7 cores for 3 second