Sunday, 19 October 2014

CPU AND GPU

A little review, focusing on CPU and GPU. Stumble upon a very interesting video, I found the sudden urge to share it in this blog.


A short video yet clarify definitions of CPU(sequential) and GPU(parallel) tasks beautifully.

Implementation of GPU and CPU together accelerate widely used applications in all platforms majorly scientific, analytics, engineering, consumer and enterprise used. CPU and GPU are dependant on each other to bring out the functions to the fullest.

HOW GPUS ACCELERATE APPLICATIONS


CPU(central processing unit), the processor or also known as the brain of the computer functions to process data which works for most common applications.
GPU(graphics processing unit) used visual-type applications which includes graphic and video.

GPUs have thousands of cores to process parallel workloads efficiently

As GPU perform tasks parallelly, it is much faster compared to CPU.
Support multithreading in which process multiple streams of data at a time. However  become a hindrance when it comes to calculation and retrieving data. When in need to fetch data from a previous process, sequential processing comes in handy. Retrieving data are focusing at one point rather than a spread search.

Source: http://www.nvidia.com/object/what-is-gpu-computing.html


No comments:

Post a Comment