Cache-Bench is a Python script that allows you to evaluate the impact of virtual memory settings (vm.swappiness, vm.watermark_scale_factor, Multigenerational Lru Framework And others) on the performance of tasks, the execution of which depends on the caching of the reader file operations in memory lack. The code is open under the CC0 license.
The main method of use is to read files from the specified directory in random order and adding them to the list until the specified Mebibyte number is. Two modes of operation:
- first – auxiliary – serves to create a directory of the specified volume. At the same time, the directory creates a specific number of metabigital files with random names.
- Second Mode – Main – File Reading Mode from the specified directory in random order. During reading, the amount of memory consumed by the script is growing, and the reading speed of a specified file value depends on the amount of cached file pages.
A part of the project is also auxiliary Drop-Cache script, which is recommended before starting the test. In the process of operation of the script in read mode, the overall operation time is displayed, the average read speed, the name of the last read file. The script also allows you to log in to a file with time stamps.