Comparison of clusters on a dedicated server and on a cloud server

	Cloud server	Dedicated server
What tasks are suitable for	For tasks with uneven load, when autoscaling of node groups is required	For load-balanced tasks and for streaming data processing
Resource Utilization	There can be multiple clients on the same host: If one client uses all the resources of a host, it affects the performance of other clients on that host; Due to the use of a single bandwidth, DDoS attacks on one client can affect other clients	Dedicated resources and isolation from other clients
Payment	Pay-as-you-go — hourly payment only for resources used	Tariff plans with prepayment for all resources ordered
Data storage	Persistent volumes (PVs) on network disks. If an error occurs on a node, the network disk will be preserved and no data will be lost	Local storage (e.g., local disks, persistent volumes (PVs) on the same server); External storage (e.g., File Storage, persistent volumes (PVs) on another server)
Auto scaling of a group of nodes	✓	✗
Changing the size of a node group	✓	✗
Changing the configuration of nodes	✓	✗
GPU configurations	✓	✗
Duration of deployment	3-5 minutes	Up to 60 minutes
Private Kube API	✓	✓
Updating the cluster certificate	✓	✓
Integration with Container Registry	✓	✓
Minor Update	✓	✗
Patch version auto-update	✓	✗
Node auto-recovery	✓	✗
Tags	✓ (creation, modification, deletion)	✓ (only creation together with a group of nodes)
Taints	✓ (creation, modification, deletion)	✓ (only creation together with a group of nodes)
User data	✓	✗
Node reboot	3-5 minutes	5-10 minutes
Reinstalling the node	✓	✗
Terraform support	✓	✗
Compliance with 152-FZ (UZ-1)	✓	✓