How GPU Computing literally saved me at work?
Python+GPU = Power, 2 Days to 20 seconds
We have all heard in some form or another the hype around GPU in recent times. Before going any further, I want to clarify, I am no GPU expert. I have just started my journey in the world of GPU. But, it’s so mighty and powerful these days and can accomplish hell lot of tasks. I was assigned a task at my work which was taking hours to run and still showed no progress. GPU came to my rescue and solved my problem within seconds. I was able to perform a task with an estimated time of 2 days in just 20 seconds.
I will state the task in details in the upcoming sections. I will also discuss how and when to use GPU for any such similar tasks. So, stay tuned and I promise it’s going to be worth the read. The basic flow will be understanding the task details, getting started with GPU and using GPU to solve the problem in hand. I will be using Python’s Numba library along with Nvidia Volta V100 16GB GPU.
I will also try my level best to keep things simple and intuitive but please feel free to jump to the references section and research about the field more. The main idea of this blog post is to help the audience understand the power of GPU and develop an intuition on how to efficiently utilize it in their daily work tasks.
