Evaluate Weigh the pros and cons of technologies, products and projects you are considering.

Helping clients with cloud assessment, decision is a critical value-add

Helping customers decide how, when and if to move workloads to the cloud is a critical value-add for partners, says David Stewart, solutions architect at 2nd Watch.

There seems to be every reason in the world to move applications into the cloud today, yet that doesn't mean you shouldn't take a rigorous approach to evaluating whether it's the best fit for your client's particular application and business. There are a few seminal factors that play into a cloud assessment and the decision to move a customer's workload to a third-party cloud service. The priority of these factors will differ from organization to organization. Culture and pre-established viewpoints will also affect decisions. The role of the value-added reseller (VAR) and implementation partner is to provide an objective filter for decision makers, and to evaluate the customer's business and IT organization for cloud readiness.

1. The application profile: All roads lead from here. When cloud computing got its start in the business world, companies thought about hosting infrastructure. As cloud computing matured, so did the options, including the acceptance of hybrid environments. Now executives should evaluate individual applications and match them to the right environment. If there are stringent regulatory requirements, or if the application will store and share proprietary information, the customer may steer away from public cloud services. Yet there's never a black-and-white scenario. Let's say a pharmaceutical company has an application for sharing publicly available human genome data and running queries on it. Given the large data sets and the high compute needs, the cloud is a more economic and practical choice for hosting this application. Yet there may be a corporate policy dictating that the analysis of the data stay within corporate walls. Therefore, the pharmaceutical company may choose a hybrid infrastructure -- cloud for the query application and on-premises for storing the analysis.

In another example, if a customer's business is subject to a standard such as Payment Card Industry Data Security Standard (PCI DSS) or a regulation such as Health Insurance Portability and Accountability Act (HIPAA), the CIO is faced with the decision of building a compliant infrastructure to deploy the application or using a third-party cloud service that still meets those requirements but will be cheaper and less troublesome than maintaining infrastructure in house. A startup with budget constraints will logically want to focus resources on its application, not on creating a compliant infrastructure. Laws and regulations rarely dictate how data is hosted -- only that certain requirements are met. As always, though, preconceptions, political viewpoints and culture often dictate the final decision.

When moving to the cloud, companies rarely have the required internal skill sets to plan, design and manage the environment.

2. Evaluate the financials: The capital expense (Capex) versus operating expense (Opex) discussion is still a big one. While the customer's CFO is probably skewed toward one viewpoint or another, partners should always advise a comprehensive financial analysis. For instance, a financial institution that must comply with the Federal Reserve Comprehensive Capital Analysis and Review (CCAR) requirement needs heavy equipment to run the numbers. Let's say the bank needs 10,000 computing cores to run the annual analysis in 72 hours. This exercise has an infrastructure profile of many servers, network infrastructure and storage. The price tag of acquiring the infrastructure alone would be large, but when combined with the ongoing operational costs of running that infrastructure along with the people costs to design and build it, the price tag could easily be far higher than most companies would want to incur for such a variable workload. Compare those costs with the month-to-month expense of running the project externally with a cloud provider and without the risk and cost of internal management and capital expenditures. For example, if the costs of the two options meet up in three years and the hardware lifecycle is three years, then the customer isn't even breaking even by doing it themselves. Also, cloud IaaS pricing wars and marketplace shifts can affect decisions. Helping customers understand vendor moves and analyst predictions can sway the decision one way or another.

3. People: When moving to the cloud, companies rarely have the required internal skill sets to plan, design and manage the environment. Senior executives will need to either hire or contract with IT experts who understand cloud architecture and can lead migration, governance and relationships with the cloud vendors. Hiring cloud experts remains elusive, given the tight market for IT pros and the reality that cloud skills are relatively brand new. In fact, the most sought after individuals with current cloud development and management skills would often rather remain consultants, given the high rates they can charge. The simpler option is to outsource those skills, a trend occurring in at least 70% of the companies that we meet. That's not to say the customer shouldn't also have a cloud champion inside, a "people person" who is adept at interfacing with vendors and consultants and overseeing strategy.

4. Innovation quotient: Speed of innovation is often paramount to the CEO and senior executive team. The cloud can be an exciting platform to support innovation, given the fact that market leaders such as AWS are releasing new services and features every few weeks. Time to market for a particular workload deployed in the cloud can be accelerated because of the nature of cloud services. Companies with an aggressive innovation trajectory will gravitate toward the cloud to grow faster and more nimbly. There are times, though, when customers have specific needs that cloud providers can't readily meet. For instance, one company our firm worked with was developing a real-time social media content sharing application that had unique "long polling" requirements for session timeouts. Their cloud partner of choice had default configurations that didn't meet the ideal long-polling times for the company's application, but the company still deployed on this provider because it believes the provider will innovate quickly enough to eventually meet these needs. This is a clear example of how a company's trust in the cloud will continue to drive innovation rather than create an obstacle around it. 

There's no question that cloud computing is changing IT organizations, probably forever. Yet there are several factors that play into what type of architecture and platform is the best choice for each customer. Staying flexible, open-minded and with an investigative approach will help customers understand the pros and cons of all the options before making the final decision.

About the author:
David Stewart is a solutions architect at 2nd Watch.

Next Steps

Survey gives insight into how customers evaluate public cloud services

More factors to guide a cloud migration strategy

Learn about cloud migration services

Quiz: Multicloud deployment knowledge 


Dig Deeper on Cloud brokering

Join the conversation


Send me notifications when other members comment.

Please create a username to comment.

What factors do you consider when helping clients make cloud decisions?
  • Need for Scalability: Not every company needs a near-infinitely scalable system. Sometimes a standard setup works just fine - that's why we call it "standard".
  • Overall Cost: Cloud services have many different pricing plans, and some are more appropriate for given needs than others.
  • Training Time: Some clients still don't have the knowledge necessary to make good use of cloud-based services.
  • Connectivity: How is the client actually going to reach and make use of cloud-based services?