How To Build A Hybrid Cloud Part 4

Finding the optimal delivery model for your organization must take financial, security, compliance and other factors into consideration.

In covering what is needed to build a hybrid cloud, so far we’ve looked at the importance of building an overall strategy before choosing products; defining business outcomes and goals; and defining capabilities and service portfolios. Now it’s time to choose delivery models.

After you have clearly identified what service capabilities you need, you can start thinking which delivery model would be best for each of those services. That’s true whether you want to build them yourself or have a provider build them for you.

The key is to find the optimal mix of hybrid service delivery model that matches meets the organization’s needs and service requirements and adheres to any financial constraints you might have.

There are several major criteria for determining the ideal delivery model for various services. One of the biggest is financial. If an enterprise is trying to reduce capital expenditures, for example, it would be best to consider vendors that provide as-a-service offerings. That way the company can scale back its internal IT infrastructure.

If regulatory compliance requirements are a big concern, such as with companies in the healthcare and financial services industries, then it’s important to use a delivery model and provider that can support compliance efforts.

The type of workloads to be supported is another factor. For large enterprise resource planning (ERP) applications, for example, companies will typically have more stringent security, performance and continuity requirements. In such cases, private clouds or a managed cloud would be the best fit. For workloads supporting applications development or public Web sites, a public cloud might be the right fit, given that these functions do not require higher performance, security and continuity characteristics.

Security is another important factor. What are the security requirements for the various workloads? If an organization is dealing with healthcare data, financial data, etc., it needs to consider solutions with features such as security information and event management, file integrity monitoring, Web application firewall, vulnerability assessment and scanning and anti-­malware software.

Another consideration is the company’s business continuity requirements, including its Recovery Time Objective and Recovery Point Objective needs. Critical workloads will require more stringent service level agreements. Here, a managed cloud environment should be considered, where providers are required to meet contractual SLAs.

Finally, there are data sovereignty issues to consider. Does the company need to keep data inside the country? Do personnel with access to data have citizenship requirements? If so, the company might need to consider a private cloud hosted in its own data center or in a provider-owned data center that meets the requirements.

Deciding whether it makes sense to use in-house delivery models or go outside the company often comes down to the above criteria. And one other thing to consider is the human factor. Does the company have the skills needed in-house to create and maintain the delivery models needed? Is it willing to forgo speed to market in order to develop the talent in-house?

Yes, there are lots of questions to ask. But the answers will help guide organizations to the most suitable delivery models for its services.


Copyright © 2015 IDG Communications, Inc.