Why do you need to be able to schedule shutdown/start cloud resources?

If you already scheduled shutdown/restart of your cloud resources, the most reason may be the cost optimization. Yes, right. But that shouldn’t be the only reason you are doing it.

How much we can save by scheduling UD (up/down) of the resources?

I need to ask back why you aren’t as it’s one of the biggest benefits when you are using the cloud. The cloud providers provide multiple methods of using the resources and payment, on-demand (the most expensive model), reserved instances, spot instances, etc. But I am quite certain that many companies are using the on-demand model because of different reasons. They have different environments (DEV, UAT, SIT, SandBox, Production, etc.) and those are not necessarily up & running on a 24x7 time basis. You may not be able to save 50% when you shut down un-computing resources, but computing instances will definitely give you immediate and linear savings if you shut down x time.

Ofc, not only compute resources will be down, if VMs are down, they are producing fewer logs, fewer network transactions, and fewer data.

This is the first reason you should consider scheduling them now.

Do you have the capability to shut down your services gracefully?

In an agile organization or distributed structure, you may have 100 microservices and 10 different types of databases. Is your company able to shut down all microservices without breaking anything and data corruption? You will have many cases & scenarios to cease your services for many reasons like DR testing, migration, cost-saving.

Remember, the computer world doesn’t even allow you to unplug your USB without the OS’s approval.

Therefore, the enterprise needs to be prepared to shut down all services gracefully based on the playbook and automated scripts.

and, vice versa. (Restart). Imagine 100 microservices (Developments) ask K8S to run their applications at the same time.

A better understanding of the Logical architecture

I believe most companies have the infrastructure diagram, application architecture, system architecture (similar but many different names). But you need to understand the data flow as well where it flows from/to. When you practice shutdown and restart, I’m 100% you will have tons of errors/logs because applications are started and send the transaction or try to connect the database. So you will know how each application data/transaction flows.

Furthermore, in very distributed infrastructure, they may not care about other integrated systems, as you assume they are up & running for your applications which is not true. You will be able to know how entire services are connected/linked like a spider web.

…Try to shut down your applications, resources.