Nearly every orchestrator has some approach to scaling either the processes its managing or the resources available to it (the cluster). Unfortunately, many of these systems are reactive. A scaling action is taken in reaction to an event that just occurred. This approach works great for unplanned events such as the lost of a cluster … Continue reading Predictive Scaling
Imagine a world where the software we build can not only react to orders, but can predict our needs and make changes autonomously. A world where those systems can learn from the past and use it to predict the future.