README.md

in_tail Scenario Benchmark

Introduction

This benchmark scenario is aimed to execute benchmark about the following target.

• tail-bench

In the above benchmark scenario, the following resources are monitored.

• CPU usage (supervisor)
• CPU usage (worker)
• RSS (supervisor)
• RSS (worker)
• VMS (supervisor)
• VMS (worker)
• Read Bytes
• Write Bytes
• Receive Bytes
• Send Bytes

Directory layout

• ansible/*
  • Ansible scripts
• config/*
  • Collection of customize Windows.
• visualize/
  • Collection of plot script from *.csv

After executing benchmark, the result is collected under ansible/output/*.

Execute benchmark

There are 3 steps to execute benchmark scenario.

• provisioning - make
• benchmarking - make tail-bench
• visualizing - make visualize

Setup

For creating instances:

$ make

Or, only creating instances:

$ make apply

And apply provisioning playbook:

$ make provision

Execute Benchmarks

$ make tail-bench

Visualization

$ make visualize

Teardown

For destroying instances:

$ make clean