Cloud asset management - Resource management

When you think of cloud computing what is the first thing that comes to mind? For me it is elasticity.  Cloud services are supposed to be designed to scale up or down based on demand.  Resource utilization rates vary from company to company, but I think we can agree they tend to be low (see Wired article for an example).

 

This low utilization rate is one of the great selling points of cloud.  Why have all the servers sitting idly “just in case”?   Why not share the hardware based on demand?  It all sounds great, especially when we keep hearing about 10-20 percent server utilization rates.  If we can create a one large pool of resources (CPU, memory, storage, etc.) won’t that help us reduce the total cost of hardware?

 

The answer is… perhaps. It probably will. Will you get to 80-90 percent resource utilization? Perhaps for a specific component—I doubt you will be able to maximize every resource.  After all, applications have different requirements so they will consume resources at different rates. You probably won’t have the perfect mix.  The high tech industry is also evolving to help address new challenges and we are starting to see new types of hardware, such as HP Moonshot System appear on the market.

 

Where are we today?

 

We have data centers (perhaps we just build a new one). We have a set capacity, which sounds very impressive on paper.  We migrate all of our applications to this data center and embark to create some cloud services.  Users flock to take advantage of the new services and our “spare” capacity runs out quickly. We modify the service definition to allow it to burst into the cloud (hybrid model).  Everyone is amazed at how great these new services are and IT gets early praise.  A year goes by and it is time to total all the costs – it turns out satisfaction is through the roof, but so are IT costs.

 

In my opinion, elasticity has to be managed. You simply cannot expand forever.  But, how can you manage elasticity?  Implementing chargeback is one great way of doing it. With chargeback, users and departments manage themselves,  making sure they stay within budget.  They will be much more likely to give capacity back once they are finished using a service to preserve their budget for future needs.

 

But, chargeback alone is not enough.  You also need to manage your existing resources. The cloud management systems like HP Cloud OS and Cloud System Automation will manage the assigned capacity, but how do you decide how many resources they should be given?   That’s where ITAM tools like our HP Asset Manager step in.  You need ITAM tools to manage resource availability, handle procurement requests and manage public cloud contracts.

 

You can read this white paper “Understand how IT asset management works for you: A best-practice gui...

 

Let’s look at a couple of examples.

 

Department A is known as a heavy user of IT.  They are quick to adopt a new service and have many active subscriptions.  In order to make sure IT budgets are not exceeded, the cloud management team limits the number of resources this department can use.  The new service is very popular and soon its consumption is approaching the resource limits.  IT is warned of this situation, reviews usage consumption and realizes they underestimated the need for CPUs.  They issue a procurement request for additional capacity. Asset management team reviews the requests and provides additional in-stock hardware to be used by the cloud service.  IT increases the resource capacity and because existing resources are used, there is no change to the price of the service and users don’t notice any service disruptions.

 

The same department continues to expand its use of cloud services.  They are within the resource utilization limits, but ITAM chargeback shows that they are approaching their budget limits.  The department manager is informed and negotiates a small budget increase.  IT updates the catalog warning users from Department A that the service availability may be restricted due to budget considerations and adds an authorization layer requiring department manager’s approval for new subscriptions.  Again, users are able to continue to consume resources, but budgetary controls are implemented in effect restricting resource consumption and forcing users to optimize use of services (decommissioning instances no longer needed in order to preserve budget).

 

Ye who controls the purse strings controls the resources

 

If we are dealing with public cloud – financial resources may in effect be the only resources that need to be tracked and managed.  In this case, the only “resource” limitations that will put in place may be budgetary.  Approval cycle may be required for new subscriptions and managers will need to track charges versus budget.

 

In summary, ITAM can play a key role in ensuring that users do not hoard IT resources and that costs stay within budget limits.  Feel free to visit our IT Asset Manager home pagefor more information. If I can answer any questions for you feel free to reach out to me in the comments section below.

Leave a Comment

We encourage you to share your comments on this post. Comments are moderated and will be reviewed
and posted as promptly as possible during regular business hours

To ensure your comment is published, be sure to follow the Community Guidelines.

Be sure to enter a unique name. You can't reuse a name that's already in use.
Be sure to enter a unique email address. You can't reuse an email address that's already in use.
Type the characters you see in the picture above.Type the words you hear.
Search
Showing results for 
Search instead for 
Do you mean 
About the Author
Featured


Follow Us
The opinions expressed above are the personal opinions of the authors, not of HP. By using this site, you accept the Terms of Use and Rules of Participation.