In this episode we will be leaving the History of Metro (which I covered in part1, part2 and part3). Let’s look at TODAY 😎

I have a lab setup ready to go and break stuff!! But first, I will walk through the lab overview (what is what) and the lab walkthrough (how is it configured). Without further ado, let’s dive in the overview! 

Oh and if you’re not into reading, you can optionally watch the SETUP VIDEO and the WALKTHROUGH VIDEO below 🙂

Lab Overview

In the lab, I have the luxury of three Datacenters (DCs): Twin DCA-DCB, and remote DCC. Each DC has its own PowerStore and several VMware ESX hosts.

Lab setup in the twin Datacenters: Each DC has a PowerStore and three ESX nodes. Zoning is local-only, meaning there is no zoning between sites; the IP connection handling the metro mirroring is the only storage-related connection which gives a really nice and clean setup.

As you can see, FC zoning is local only; there is no FC leaving the datacenter. The only storage-related connectivity is the IP connectivity between the PowerStores. This connection builds and maintains the Active/Active metro mirroring.

Site DCC (the third site) contains all the management stuff. It has its own PowerStore, and two ESX nodes. It runs things like the vCenter (no not cloud what happens if we kill resources on DCA or DCB) and it can also hold the witness VM:

The complete three-site setup. Twin DCA and DCB have an active/active metro configuration and site C carries supporting VMs such as vCenter and the optional Witness VM.

All that we need to add now, is the vSphere configuration that we run on top of the ESX hosts:

The complete lab setup. DCA and DCB have a stretched vSphere cluster configuration and there is a smaller management cluster in DCC. 

Spanning across DCA and DCB I configured a stretched cluster. This cluster has a full metro configuration. DCC just delivers the supporting VMs such as vCenter and the optional Witness VM. 

Workloads running are four Windows VMs; Win1 and Win2 are running in DCA and DCB respectively, and they are running on a preferred volume. Win3 and Win4 are the other way round; they run on non-preferred volumes. As we break stuff we should be able to observe a difference!

As with the entire series, you can optionally watch this story as a video as well 🙂



If you want to see how this architecture was actually implemented in the lab take a look at the second video:

Leave a ReplyCancel reply