The cloud market, both public and private, is still changing. Although it has definite industry leaders (AWS, Microsoft, IBM, Google and Salesforce), a multitude of providers are in the marketplace today, each offering unique services. The way people interact with technology is evolving at an unprecedented pace. Other than agreeing that the cloud market will experience massive growth globally, predicting what that market will look like, or what new features will dominate in five years, is difficult.
One attractive aspect of public cloud is that it gives companies the ability to provision services without long-term commitment. But this perspective runs against the norms familiar to most CIOs. Unlike traditional IT, where capital expenditures drive decisions favoring existing vendors, the cloud encourages companies to shop for the best resource provider for each workload. Provisioning decisions now involve capacity needs driven by seasons or cycles, not maximum service levels required over years. Things like spot markets where you can trade capacity a cup at a time were unheard of in the past. Flexibility makes the cloud attractive and, by default, encourages companies to change vendors to find the best fit. Therefore, a certain amount of churn in your provisioning strategy should not be seen as an indication of failure.
Here are some factors contributing to companies changing providers:
1. Changes in available cloud providers. The cloud market, like any emerging market, is undergoing constant change. A number of new entrants are in the mix, each with unique pricing plans and service offerings. A number of CSPs have been absorbed by others, and a few have gone out of business. The market is still consolidating as technology evolves. Mobile apps, the Internet of Things and big data have all descended on, or you could say risen to, the cloud.
2. Cost savings. Many companies embraced the cloud on the basis of early stories of cost savings. In many cases, savings were difficult to prove. Often there was a poor understanding of the cost of the internal infrastructures being replaced. This situation did not mean savings were absent but that the companies were unable to provide compelling evidence of savings.
It was also hard to buy “like” services. Standard offerings from cloud providers often included more features and functions than an individual company provided. CSPs most likely had more service coverage, more highly skilled technicians, more backup and recovery options, more security, and more-frequent software upgrades. When considering the costs of these features while evaluating savings, the CSP would often show greater savings.
3. Shadow IT. In many cases, the CIO was thrust into managing cloud services inherited from contracts negotiated by another department. These one-off selections may have been highly successful in terms of the organization that made the sourcing decision. But once identified, those shadow arrangements needed to be assessed in terms of the corporation’s overall cloud strategy. Sometimes better alternatives were found. Sometimes the better choice was to integrate the workload back into existing applications.
4. Economies of scale. Using a public cloud allowed companies to embark on trials of new technologies and processes without making long-term financial commitments. Some companies deployed smaller workloads as a proof of concept. As the workloads gained critical mass, it became cost effective to invest in internal resources, both hardware and personnel. Hence, a move away from the public provider.
5. Spread of SaaS. Software-as-a-service offerings in the public cloud add agility to meet changing market needs. The days of long application-development cycles have been left behind in favor of ready-to-use applications. Entities that use internally developed applications deployed on company-owned infrastructure or that use IaaS or PaaS in the cloud have chosen to move work to providers offering full SaaS solutions.
6. Complexity of hybrid deployments. IT organizations have learned to create complex architectures that blend public and private cloud resources, exploiting the best features of each. Some have used public providers for processing power while maintaining all data stores internally for security reasons. Others have made extensive use of public-storage options, especially for DR, but continue to process data on internal resources. Successful combinations like these have encouraged IT organizations to explore more-complex options and have contributed to changes in where and how workloads are provisioned.
Early cloud usage was dominated by a desire to reduce operating costs. Today, companies have learned more about evaluating workload-placement strategies considering scalability, agility and flexibility in addition to costs to determine which workloads are best suited to a public provider environment, as well as which providers are the best match. Astute cost-management practices recognize the true cost of moving workloads from one vendor to another. Many times, the longer-term benefits of moving to a new provider outweigh the cost of repositioning the work. Change does not equate to failure; it equates to learning how the cloud infrastructure works and learning to exploit this emerging technology to its greatest advantage.
About the Author
Penny Collen is the Financial Solutions Architect for software provider Cloud Cruiser, Inc, where she enables organizations to successfully move to hybrid cloud environments. Her role includes providing a foundation for optimizing cloud investments, enhancing IT services for the business and developing activity-based cost models for hybrid IT. Penny is an industry expert and thought-leader on the management of IT with special emphasis on financials.