Episode 47

Big data analytics is generally answering questions about the past, whereas streaming analytics answers questions about the present. But what if you could bring the two together and answer questions about the present based on what happened in the past?

Let’s use a simple one-dimensional example to illustrate the utility. Is your car overheating or not? Streaming analytics would compare the current engine temperature with a pre-defined maximum value. As long as the engine temperature isn’t above the max, all is good. However, let’s now say that maximum value changes depending on where your car is, the outside temperature and the speed in which you are going. (Big) data analytics, based on the three variables would supply a constantly changing maximum value – maybe not needed for the minivan but it could be of value to the F1 race car being pushed to the limits.

Now this example is in 1-D. Think about the complexity of the problems that could be solved, beyond overheating, when looking at two values (2-D), or three values (3-D) and higher? This one-two punch type of analytics can’t be handled by all database/application architectures but new ones that can are starting to emerge.

Here’s What We’ll Cover in this Video:

Watch this video to see Eric Tran-Le discuss a state of the art analytics.

Mentioned in this Episode and Other Useful Links