I am a big believer in collecting data an an APPROPRIATE scan rate and storing an APPRORIATE amount of it. Some processes (like the one shown) are slow moving, and there are certainly very fast processes that require the type of speed/fidelity shown here. But this is ridiculous.
Here is Part 2 of my RANT, where we almost missed an instrument failing:
Good rant Jim. Data fidelity is one aspect of data quality, a subject that seems to be getting a lot more attention recently. Gopal, at OSIsoft recently made a blog post on A Layered Approach to Data Quality. There will be a series of webinars on the subject. https://pisquare.osisoft.com/community/all-things-pi/blog/2020/07/11/industrial-iot-time-series-data-engineering-a-layered-approach-to-data-quality