Here i will let you know all about DRS
1. What is DRS
2. DRS Resources management capabilities.
c.Power Management (DPM)
3. DRS Pre-requisites
4. DRS Automation Level
5. Using HA and DRS together
Lets start with
Q. What is Distributed Resources Scheduler (DRS)
DRS is enabled in a cluster. A DRS cluster is managed by vCenter Server and has these resources management capabilities.
- ) Initial Placement
- ) Load Balancing
- ) Power Management (DPM)
Lets discuss about these above point.
- Initial Placement:- when you first Power ON virtual machine in the cluster, DRS either places the VM on an appropriate host or sends a recommendation.
- Load Balancing:- DRS continuously monitors the distribution and usage of CPU and Memory resources for all Hosts (ESXi) and VMs in the cluster. DRS compares the resource usage defined by the migration threshold and then performs or recommends for VMs migration accordingly.
- Power Management (DPM):- when DPM is enabled in a DRS cluster, DRS compares the cluster level and hosts level capacity to the demands of the clusters virtual machines. if a host1 which is underutilized, evacuate all the VMs to some other host2 and host1 made powered off or standby power mode. once the demand or usage grows the VMs distributed across the hosts on the fly without any downtime and the standby host will be powered back ON.
Note : Generally we do not use this DPM option in Production environment.
DRS Cluster Pre-requisites:-
- DRS works with vMotion so if the VMs meet vMotion requirements this would be best.
- To use DRS for load balancing the hosts in a cluster must be in a vMotion network.
- Hosts must use shared storage (VMFS data store or NFS data store)
- Place the disks on a shared storage that is accessible by source and destination hosts.
- Ensure that the shared storage must have sufficient space to store all virtual disks for your VMs.
DRS Cluster Automation Level:-
- Manual:- When you power on a VM, DRS displays a list of recommended hosts on which to place the VMs. when the cluster becomes unbalanced, DRS displays recommendations for virtual machine migration.
- Partially Automated:- When you powered on the VM, DRS auto place the VMs to the best suited host. When cluster becomes unbalanced, DRS displays recommendations for VM migration.
- Fully Automated:- When you power on the VM, DRS auto place the VMs to the best suited host. when cluster becomes unbalanced, DRS migrates the VMs from over utilized to underutilized hosts to ensure a balanced use of cluster resources.
Using HA and DRS Together:-
- HA closely intergraded with DRS. When a failure occurs, HA first checks whether resources are available on that destination host for the failover. if the resources are not available then HA asks DRS to accommodate for these resources. For Example: A VM that has been assigned a large CPU and memory reservation that results in fragmentation resources i.e. inefficient use of resources throughout the cluster. vSphere HA requests but can’t be guaranteed, defragmentation of resources to accommodate for the VMS resource requirements.
- DRS flatters shares and limits on virtual machine before failover. This flattening process ensures that VMs get the resources that they are entitled to if they would have been failed over to the correct resource pool cause in the earlier versions the VMs failed over to the Root Resources Pool even if it is available in the other child resources pools.