26 Jun 2022
Resources:
vSphere High Availablity (HA) allows VMs, in the event of a failure, the virtual machines on a failed host are restarted on alternate hosts.
VMware vSphere Distributed Resource Scheduler (DRS) is a feature included in the vSphere Enterprise Plus.
Requirements of DRS:
Before vSphere 7, the focus was on the cluster and the ESXi hosts, ensuring that hosts are evenly balanced with their VMs and workloads.
After vSphere 7, the focus is now on the workloads of individual VMs. Each VM is calculated a score out of 100 (higher is better) based upon the following:
vSphere 7 - Improved DRS blog article, below is a walkthrough GIF of DRS.
DRS can be set up in a cluster either during or after initial setup.
DRS can be configured to be in one of the following modes:
Predictive DRS uses a combination of DRS and vRealize Operations Manager to predict future demand and determine when and where hot spots will occur.
Virtual Machine Automation allows for specific VM migration configuration for example vCenter.
VM Distribution allows for even distribution of VMs across all hosts within a cluster. low intensity workloads may end up being grouped and if a failure occurs the fall out could be massive, even distribution attempts to reduce this.
CPU Over-Commitment sets a limit of virtual cores against physical cores. This rule does not apply during times of high availability.
Scalable Shares are new to vSphere 7, this topic was discuss on the Resource Management article.
Distributed Power Management (DPM) will consolidate VMs into a set of hosts and power off hosts that are not required. When the workload increases the offline hosts will be powered on and vMotion VMs.
IPMI or iLO Settings need to be configure along side Wake-on-LAN (WoK) to utilise DPM.
The Advanced Options tab allows for custom parameters.
Monitoring is performed under the vSphere cluster using the Monitor tab and vSphere DRS.
Navigate to Cluster > Configure > Configuration.
Groups can be created to group either VMs or Hosts together and name them, these can then be used in other rules.
Concepts include:
Rules can then be created to restrict Development VMs to Development Hosts.
Navigate to Cluster > Configure > Configuration.
New rules can be created to perform tasks such as:
Navigate to Cluster > Configure > Configuration.
VM Overrides allows for specific VMs to be configured to override global / cluster settings such as DRS automation levels, restart priorities etc.
When entering maintenance mode there will be a prompt for moving powered-off and suspended VMs, this is the first step when decommissioning an ESXi host, otherwise the VMs will become unregistered.
If the DRS cluster is in manual or partial only DRS recommendations will be generated, if the cluster is in fully automatic mode the VMs will be migrated using vMotion and the host will enter maintenance mode.