An In-memory Key-Value Store and Load Generator

One Paragraph of project description goes here

Getting Started

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.

Prerequisites

Clone the repository to get a copy for yourself

$ git clone https://github.com/mohapeameya/kv-store-load-gen.git

Change current directory to kv-store-load-gen

$ cd kv-store-load-gen

Install the following dependencies if not already installed

$ sudo apt install g++ -y
$ sudo apt install util-linux -y
$ sudo apt install python3 python-tk -y
$ pip3 install matplotlib

Installing

Run make all to build

$ make all

Running the test

For valid test results we need to ensure the following:

1. There is a clear separation between server and load generator resources
2. Load generator can saturate server without becoming a bottleneck itself

This can be achieved using taskset utility by pinning the processes to specific cores
For example, in a 4 core system, run server locally(127.0.0.1) on port 55555 with 5 worker threads on core 0

$ taskset 0x1 ./server 127.0.0.1 55555 5

and loadgenerator on cores 1,2, and 3 with minimum 1 thread and maximum 5 threads with 10 seconds per thread

$ taskset 0xE ./loadgenerator 127.0.0.1 55555 1 5 10

Graphical visualization of test data(wait for the load generator to exit)

$ python3 ./plot.py

Authors

• Ameya Mohape - Initial work - mohapeameya

License

This project is licensed under the GNU GENERAL PUBLIC LICENSE Version 3 - see the LICENSE file for details