Typically large complex problems that demand a big system to do it. Let me give you a real example. Imagine you are trying to improve the life chances of a premature baby in an incubator. You might need to collect over 1000 unique pieces of information every second from a range of sensors. You need to transfer all that data to a computer that can process all of that information in real time and then make a recommendation to a doctor. Actually that isn’t a computing problem. it’s more a data movement problem so you need high speed, low latency networks to do that and then a large parallel computing setup to actually process all that data as it arrives. The end result is something that moves the data, does the calculations and delivers better life chances as a result.
Comments