Previous | Next --- Slide 4 of 46
Back to Lecture Thumbnails

The difference between a supercomputer versus a data center is that data centers serve millions of customers at once, crunching data, and organizing/computing data + results.


Really cool link for folks who are interested in what happens in a Google data center and what it actually looks like and how all the pieces come together. I would recommend watching the video for a clear, but concise breakdown and visualization!


Out of curiosity, don't websites such as Google that service billions of requests each day have supercomputers to process them?


TO add on to @efficiens, does this mean that even if there is a huge amount of data, as long as there are enough servers, we do not need supercomputers to process the data?


@althalus, @efficient Super computer is designed for compute-intensive applications, while data center is designed for data-intensive applications. For most applications, the computing capacity of super computer is not necessary.


The two main large-scale systems are supercomputers and datacenters, each with a different focus. It would be ideal to have a computing system with both high data intensity and computational intensity. However, there are challenges at hardware, runtime and application levels.