With the new release with vCetner 7, there is a new and improved Distributed Resource Scheduler ( DRS ) and in this post, we will go through some comparison with the original DRS and the new DRS.

In the Orignal DRS which has been there since VMware introduced, no doubt there has been enhancements and additions to it example we only had Compute DRS later VMware added Storage DRS leveraging vMotion and Storage vMotion of course

Orginal DRS Cluster centric

Runs every 5 minutes

Uses Cluster -wide Standard

Uses Cluster-wide Standard deviation model

VMware has done great thinking about how the new DRS 2.0 should work and they have changed the algorithm to work on the modern application.

The New DRS is Workload Centric so instantly we get better performance

Improved DRS

Workload Centric

Runs every 1 minute

Use the VM DRS score based on granted memory (New Matrix)

Let’s discuss how this new DRS works a little bit. VMware has started using a new matrix for workload, or VM. its called VMDRS Score

Improved DRS using VM DRS Score

VM DRS is necessary because with the modern-day application we need to focus on the VM / Workload perform as appose to cluster performance which the Orignal DRS cared about.

so we have a fundamental change on the way the improved DRS works using VM DRS score which is the foundation of the DRS. lets a closer look at it.

we already explained that the new improved DRS runs every 1 minute.

We have in the UI a lower bucket score from 0 to 20 % or 20 to 40% whenever we see another host can give a lower bucket score the DRS considers moving the VM.

the vm DRS score is calculated using a lot of matrices. example CPU ready time, capacity, headroom of ESXi host. network utilization.

Finally, it also checks the migration cost of the VM. ie how much network resource is going to be used in doing this migration if its going to expensive or performance constrains DRS migration will not take place.

To learn more about new features with vSphere 7 and vCenter 7 click on the links below:

VMware Cloud Foundation 4

vCenter Server Profiles

Content Library with version control

New and improved DRS 2.0

vSphere 7 and vSGX Intel Software Guard Extension

vCenter Server Upgrade Planner

Improved vMotion

What is DRS?

VMware vSphere® Distributed Resource Scheduler (DRS) is the resource scheduling and load balancing solution for vSphere. DRS works on a cluster of ESXi hosts and provides resource management capabilities like load balancing and virtual machine (VM) placement.

Benefits of DRS?

VMware runs its own algorithm in the background to balance the Cluster resource, primarily matrix that is taken into account are CPU and Memory utilization on the ESXi host.

However, from the version 6.7 VMware is now going to DRS based on network utilization too this is a feature of DRS.

How to enable DRS?

To enable DRS we need to right-click cluster and edit settings and select DRS

You can set the DRS automation level in the client from Cluster > Manage > Settings > Edit Cluster Settings

How to configure DRS Options.

We have 3 types of options once we enabled DRS.

Manual

VMware will not do any automatic DRS it will be down to VMware Administrator to make that decision after checking the host resources. You must apply both initial placement and load balancing recommendations

Partially automated

VMware will recommend, some DRS and the VMware admin can then accept them which will be applied. DRS applies recommendations only for initial placement ( usually, when power on a VM VMware can suggest to move it to a host which is underutilized. )

Fully automated

In fully automated we have 5 options from 1 which is conservative to 5 aggressive. DRS applies both initial placements and load-balancing recommendations automatically.

Just below once we select our cluster to be Fully automated we will find this amazing slider bar which lets you control the movement of your VM’s from conservative to aggressive as shown below.

1 DRS will only apply recommendations that must be taken to satisfy cluster constraints like affinity rules and host maintenance. DRS will not try to correct host imbalance at this threshold.

2 DRS only provides recommendations when workloads are extremely imbalanced or virtual machine demand is not being satisfied with the current host.

3 DRS provides recommendations when workloads are moderately imbalanced. This threshold is suggested for environments with stable workloads. (Default)

4 DRS provides recommendations when workloads are fairly imbalanced. This threshold is suggested for environments with bursty workloads.

5 DRS provides recommendations when workloads are even slightly imbalanced and marginal improvement may be realized. For dynamic workloads, this may generate frequent vMotion recommendations.

Predictive DRS:

With a simple click to enable predictive DRS, you can now allow VMware to act upon predictive hardware failures. The ESXi host will be put on maintenance mode which will avoid hos failure situation and outage to you very loved production or development servers.

VM Overrides:

In some cases, you might require DRS to treat some VMs as special. For example, you might decide DRS should not consider a specific VM when generating its recommendations, or you might decide DRS should not migrate that VM at all. So for these special VM you can set VM Overrides.

You can set VM overrides under Cluster > Manage > Settings > VM Overrides,

as shown below you can set the automation or migration threshold for a VM to a value different than that at the cluster level, or even disable them.

Options available are Disable, Manual, Partially Automated and Fully Automated.

I hope you enjoy reading this post, you can find a quick video on how to configure in this video below:

References: httpss://www.vmware.com/content/dam/digitalmarketing/vmware/en/pdf/techpaper/vsphere6-drs-perf.pdf

Leave a comment with your response.

Don’t forget to #subscribe Enjoy watching and keep #sharing!