Why in the third decade of the 21st century, after 80 years of writing software, data continues to live in two states, hot data in the compute space and cold data in the persistence state?
Of course, data is created and manipulated while in a hot state. Ideally, all data would exist entirely in a hot state when you think about it. However, the reality is that due to the way most computers work, it is necessary to keep only the data that is actively in use in a hot state and push all inactive data out to a cold state persistence layer.
This hot and cold data perspective raises the question of why is it that at the software level, we still have to focus so much attention and effort on dealing with the movement of data between these two states.
In this talk, we will examine what has led us to this point, why we seem to be stuck here and look at some possibilities for moving forward. You also see a real example of a live hot state software approach that inspired this rant and line of thought about seeing data in a hot and cold state way and the profound impact on how we develop software.