Using other mathematical models in observability
SREs can go beyond traditional statistical methods by either applying computer-based simulation or artificial intelligence to extract augmented knowledge from monitoring data points. Computer-based simulation can run uncountable scenarios against a statistical model to forecast results. Machine learning employs dataset modeling to identify patterns usually hidden inside them. Let’s talk about these two examples next.
The Monte Carlo simulation
This is an algorithm that simulates an event with randomly generated inputs, calculates the output based on a model, and repeats this cycle many times. The Monte Carlo algorithm estimates the probability of an event happening after thousands of simulations with random variables. For instance, we could create a model of an app to estimate the probability of a component taking more than 2 seconds to respond after 10,000 simulations with random values as input. We can use Python with...
