- Category: The Basics
- Published: Thursday, 21 February 2013 18:43
- Written by Zack Kielich
- Hits: 27126
VMware vCloud Automation Center, or vCAC, is an enterprise tool to allow end users the ability to consume your organization's cloud while giving cloud admins the tools to curtail the consumption. The idea is that an end user naturally shouldn't have access to the hypervisor manager and at the same time doesn't understand enough about virtual infrastructure to use a product like vCloud Director (if you're building a Software Defined Data Center). This leaves cloud admins with having to do nothing but field requests to create, destroy (like that ever happens), or change virtual machines.
vCAC seeks to fix this by displaying items specific to the user's access in a catalog. End users are then able to make an order and have the needed infrastructure built for them through a process that a cloud admin has approved. No longer do you have to pull an entire cloud admin team off of what they're working on to fulfill a 50 VM request.
Faster VM creation:
Since all of this is automated, it can complete the requests faster and in a standard way-- meaning that there's no more 2 week turn around or variations in configurations. In addition, because customers are getting what they requested in fairly short order, it gives them less reason to go around your processes by putting a their corporate credit card down at a public cloud vendor. This isn't to say that those public cloud vendors can't be part of the cloud that your cloud architects design, but you now have control over which workloads go to the public cloud (for security) as well as how long they stay out there (to manage cost). The quick turn around also means that they don't try to horde machines that they aren't using, for fear that it will take forever for them to get another if they need it.
Since the end users can spin up the machines quickly they don't mind as much when a machine is de-provisioned after a chosen period of time. When the machine de-provisions at the end of the lease, it follows a process to return the resources back into the pool of available resources. That means no more "unscheduled SAN outages" to turn off all the virtual machines to see who screams, indicating that the machine is actually being used. Also, you no longer have to keep ordering hardware to feed VM-sprawl.