Recommendation for VRA Infrastructure frabic ?

[Rajisom]

Please,your suggestion for VRA Infrastructure fabric and Fabric group creation in following scenario.

Total 4 DC ,VRA in primary DC, Total 3 tenant for single Organisation.

1.What is the best practice to create the Infrastructure Fabric ? seperate infrastructure fabric for each Site ? or Single fabric, consolidating endpoints from all 4 sites ? ==> please check the DR scenario -> VRA fabric during VRA

[Rajisom]

How many !aaS fabric is possible in a VRA ?

[daphnissov]

We would need more information from you as to how vRA is setup and required to work in your environment.

[Rajisom]

What exactly ?

[daphnissov]

1. You said you have one organization but three tenants. What is your use case for separate tenants rather than using business groups? Because now you've tripled your work since these groups have to be maintained separately in three different locations.
2. How many clusters do you deploy to in each of the four data centers?
3. Are there different people/teams responsible for carving up resources originating at each of the four DCs?
4. How often are you carving out resources once the initial reservations are created?

There are others, but I'm not going to list everything. What you essentially have to ask yourself is "who is in control of resources to be divvied out to vRA for consumption?" If that's only one person/team, then a single fabric group will do. If that's separate teams/users, maybe per DC, then you'll need to create multiple fabric groups.

[Rajisom]

Thank you for your time !

1. The customer is hosting three environments under the VRA ...Pre-Prod. Dev and Prod Environment.They got the different set of users who will be making a request for provisioning.Hence they want complete isolation between prod and a Pre-Prod environment.

2.there is going to be minimum 16 clusters in each site with each cluster containing 16 nodes.

3.This will be a common team who will be doing the carving up

4.This is a huge migration project and there are high chances that regular carving up is happening.

In my solution, the VRA components will be hosted in primary DC and all the VMS/appliances will be replicated to DR site via SRM. With this approach, during the DR, the primary site will be lost and all the VMs will be failed over to DR.

That time, the end points in primary site won't be available.Will that make any issue to the recovery of the Fabric and Fabric Groups?

I have another query also.

What is the difference between the fabric models "'Multitenant with Infrastructure Configuration in Each Tenant  /Multitenant with Infrastructure Configuration Only in Default Tenant"?

Is there any VMWAREW documentation which describes the pros and cons for the same.

[daphnissov]

I'd recommend that you re-think your decision to host individual tenants for these groups as it is going to complicate things. In my experience, these are better served as business groups that can be isolated, and very infrequently do setups like what you describe call for separate tenants. It sounds like it may be better to have one fabric group and put all resources under their control. Having a single tenant will make that simpler. Endpoints and fabric groups are unrelated in terms of functionality if attached resources are not available. As I believe I said in your other post, your focus needs to be on controlling reservations and blueprints with the forethought of an entire site being unavailable. The general vRA documentation site describes your question as far as infrastructure configuration is concerned.

[Rajisom]

https://docs.vmware.com/en/vRealize-Automation/7.0/com.vmware.vrealize.automation.doc/GUID-12E7F3C7-...