Autoscaling of Loki components to address short term spikes

Proposed title of this feature request
Autoscaling of Loki components to address short term spikes

What is the nature and description of the request?
Overall resource usage of OpenShift Container Platform 4 - Cluster Logging with LokiStack is already fairly intense. Yet in many situation a smaller deployment size may be sufficient as only during specific times, number of logs and thus load may be elevated. Given the auto-scaling capabilities, OpenShift Container Platform 4 does provide, it's requested to leverage those capabilities with OpenShift Container Platform 4 - Cluster Logging `LokiStack` as well and increase number of replicas during peak times and lower it to the bare minimum during regular times or not business hours.

Why does the customer need this? (List the business requirements)
Cost of resources are a constant concern. If some peak time demand to deploy OpenShift Container Platform 4 - Cluster Logging LokiStack with a bigger deployment size to just cope with peaks it's a waste of money and resources. So if a smaller deployment size could be configured but have the same scale out during peak hours, once could safe a lot of resources and money and also nicely demonstrate the capabilities and benefit of the overall OpenShift Container Platform platform.

Also, considering that many environment will have multiple OpenShift Container Platform 4 Cluster and therefore multiple OpenShift Container Platform 4 - Cluster Logging stacks, there is a lot of money involved that can be saved and used otherwise. It's also an economical point that needs to be considered, when only the resources need to be made available that are actually required and not have to size everything for potential peak times.

List any affected packages or components.
OpenShift Container Platform 4 - Cluster Logging LokiStack