From an episode of Byte Into ITPresented by Dan Salmon and Row Murray


Byte Into IT: Conrad Sanderson on GPUs (Graphic Processing Units)

CSIRO scientist Conrad Sanderson joins Dan and Row to explain 'GPUs', or graphics processing units. They were originally designed for rendering complex graphics, but in recent years have become integral for and enabled the boom of AI due to their ability to process large amounts of data "in parallel" (at the same time), unlike some chips (such as CPUs) which process data sequentially.

Machine learning techniques in AI, such as deep neural networks, rely on various forms of 'matrix multiplication' (a mathematical operation where very large sets of numbers are multiplied and summed together). These operations are well suited to parallel processing and can be performed very quickly by GPUs. Its horsepower is why some companies who manufacter top-of-the-line GPUs are worth billions.

Listen to Byte Into IT: Conrad Sanderson on GPUs (Graphic Processing Units)10:461 May 2024