Cloud Management Platform (CMP) – Intelligent Provisioning and Optimization

Cloud Management Platform (CMP) – Intelligent Provisioning and Optimization

This post was originally published on this site ---

One of the main drivers of a Cloud Management Platform (CMP) is to have the different solutions work together seamlessly.  That means the business, automation, compliance, operations, optimization, capacity, etc…  All these components need to work to provide you with a unified feel and flow.  Without that its really not much of a CMP…  its just a bunch of features and functions cobbled together.

One such integration point that is critical to any CMP, and to your business as a whole, is that of your automation and operations tools.  For VMware that means vRealize Automation (vRA) and vRealize Operations (vROps).  While these two solutions have always delivered better value together, the latest releases of vRA 7.5 and vROps 7.0 have brought major enhancements to the VMware CMP solution.

Initial Workload Placement

The ability to properly provision new workloads into your environment is of upmost importance.  Placing it incorrectly could lead to all sorts of issues for compliance, SLAs, application performance, cost etc.  It also means you will just need to manually move the VM later!  The initial workload placement process between vRA and vROps takes care of all of this for you.

vRA provides the governance (infrastructure privileges – what users have access to what part of your infrastructure) and business intent (SLAs, license compliance, compliance and cost).  vROps on the other hand is responsible for the operational intent (densification, balance, headroom and application performance).  Together they form a powerful solution for deploying your workloads.

  1. It all starts in the vRA customer portal where a user can deploy new workloads and applications from the self-service catalog. vRA’s reservation policies dictate where the workload(s) can be placed in the environment based on the blueprint and the user’s privileges.  Does it need to run on PCI compliant clusters?  Should it run on the fastest storage?  Should it be placed in the test/dev datacenter?
  2. The list of possible places (e.g. datacenter or custom datacenter) to deploy the workloads is sent to vROps which holds your operational intent.
  3. If the deployment is for business-critical apps where performance is paramount, then a balanced approach is recommended. But if the workloads to be deployed are for testing, then maybe a more consolidated approach is warranted.  vROps also ensures that any headroom settings you may have implemented, to lower the possibility of contention from sudden resource spikes, are honored.  Based on your configured operational intent vROps determines the BEST place to deploy these workloads (e.g. cluster).
  4. The placement recommendation is then sent back to vRA.
  5. vRA deploys the workload to the correct cluster. DRS will then place the workloads on the appropriate host(s) within the cluster.

DONE!  Simple and sweet.


Ongoing Workload Optimization

As we’ve discussed, correctly placing workloads in their initial destination is important.  Of course, we hope it stays there throughout its lifetime and never needs to move anywhere, but let’s be honest that’s never the case.  Applications change, workloads are added or deleted, HW is refreshed and all of this can affect the health of your workloads/applications.

Together vRA and vROps provide automated workload optimization functionality that will ensure the business and operational intent of your workloads are met throughout the lifecycle of the applications.

  1. vROps is continuously looking for optimization opportunities based on your operational intent. Is a cluster experiencing contention?  Are there places where we can consolidate?  Are any of the clusters breaching my configured headroom setting?  If the answer to any of these is “yes” vROps will start a workload optimization process to resolve it.
  2. Before recommending any moves vROps needs to understand the business intent of the workloads. To do this it grabs the reservation policies from vRA which tell vROps where workloads CAN be placed.  With the knowledge of the current state of the infrastructure and your business needs, vROps formulates a workload optimization placement plan.
  3. This plan is broken down into the actual moves that need to be made to bring the environment back to an operational and business “green state”.
  4. The list of moves is sent to vRA.
  5. vRA executes the plan thereby ensuring BOTH products in the solution are aware of the changes.


Together vRA and vROps will help you drive the BEST possible value out of your environment, keep your applications happy and healthy, help lower your overall costs, drive compliance and governance and just make your job easier overall.

Want to see it all in action?  Check out this video.

The post Cloud Management Platform (CMP) – Intelligent Provisioning and Optimization appeared first on VMware Cloud Management.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.