Efficient Update of Indicators

[JSunRae]

Hi all, I was just hoping to find out what ideas everyone has for updating indicators efficiently as bars update. I have built a good system that I am happy with, and backtested. Unfortunately I am now working on live testing and not found a way to streamline the flow of information efficiently.

Currently I have a dataframe holding all the bar information (ticks, or high/low/close/volume). Another dataframe with the technical indicators calculated. Which unfortunately isn't very efficient as it recalculates the last 200 entries for every new bar. Then because I am using ML; a third dataframe with everything normalised. Its really a mess.

Just wondering if anyone wants to share any tricks or tips they have. If I solve it in the next few weeks I'll update with my attempt.