According to a January 2018 Forbes article, 83% of enterprise workloads will be in the cloud by 2020—but how many of these organizations rushing to migrate their data fully understand each cloud service model? More importantly, how many know which models are the best fit for their organizational needs? Also, what must an organization consider when selecting the best models for their overall cloud strategy? Let’s begin by understanding the features, advantages, and disadvantages of the three cloud service models.
Infrastructure as a Service (IaaS)- With IaaS you’re buying the compute, storage and networking from a provider—acquiring the infrastructure support while maintaining a high degree of control and freedom to run your business.
All levels of compute can be provided—from virtual machines and full server hosts to just operating systems or containers. Usable disks of specified size provide storage capacity based on individual preference. Networking connectivity with firewalls and security are typically standard with these packages. All provisions with IaaS are available in different quantities, based on individual needs and are frequently billed as “Pay-As-You-Go” by usage.
The main advantage of IaaS is the freedom and capacity to configure and control all applications, compute and storage without the hassles of provisioning or managing the infrastructure. Your enterprise receives the keys to the capacity, retaining complete control over business functions. This dynamic freedom can also act as a disadvantage—more freedom can bring significantly more management work.
Platform as a Service (PaaS)- With PaaS, you’re buying the preconfigured infrastructure required to directly deploy applications including all the libraries, services, toolsets, and programming language components. PaaS provides the keys to a fully-fledged, built out development platform environment.
The main advantage of PaaS is the minimal management required for this preconfigured package. PaaS removes many of the concerns and provisions mandated under IaaS—allowing developers to configure applications with minimum delay.
The disadvantages of PaaS include less organizational control and reduced ability to customize. Teams must adapt to the limitations of an out of the box, one-size-fits-all configuration.
Software as a Service (SaaS)- Considered the platinum, top-level offering by many, this is also the most familiar model. With SaaS, you’re purchasing the infrastructure, platform and applications all on the cloud and accessible from a network or device. SaaS only allows the user to configure the application and includes no control of infrastructure or platform—merely consumption of the software.
The advantage of SaaS is by far the least work and management—simply plug in and go, freeing maximum time to focus on business. The primary disadvantage of SaaS is the least organizational control of the three cloud service models.
CONSIDERATIONS IN MODEL SELECTION
Every enterprise must calculate how much of each service model they want to incorporate in their plan to maximize the benefits of the cloud. The perfect strategy balances the advantages of each model with the needs of the organization. When selecting the appropriate service model, an enterprise must consider:
End Applications- Start by examining your end applications and work backward, seeking the models that best fit. For every application consider these important questions.
- Where do I want to run it?
- How much do I want to own?
- How much do I want to control?
- How much can be run as a service?
- How much can be left to vendors or package software?
Level of Control- Assess your comfort level for the amount of influence your organization will have over infrastructure, platforms and applications. Is it critical to your business to control and manage these or do you view them as commodity functions someone else can manage, and you merely access based on need?
Amount of Work- Determine how much sweat equity your enterprise wants to allocate into the configuration and management of its infrastructure, platforms, and applications. Service models allowing more control also require more work. Is your enterprise willing to shoulder the extra management burden in exchange for a heightened directive?
Cost- In cost terms, there is always an argument for taking on more in-house. However, increased autonomy brings care and feeding—including the added employee costs associated with management. Also, selecting a service model based primarily on cost also overlooks the real value of the cloud—freeing time and resources to focus on growing your business.
THE TIME IS NOW
The best news for all enterprises still struggling to select the best service model and formulate their overall cloud strategy is you’re right on time—and not alone. The great myth is this mass migration to the cloud means companies have all this figured out. In reality, only 8% of respondents answered yes when asked if they had an enterprise-wide, formally established, published cloud strategy according to a survey from “Gartner Essentials: How To Create an Enterprise Cloud Strategy” by Bob Gill, presented at the Gartner IT Infrastructure, Operations Management & Data Center Conference 2017.
Kevin Westendorf, CTO and VP of Technology for the Cincinnati-based cloud provider CenterGrid expounds on this theme. “If you’re still struggling to figure out the best cloud strategy that balances the ideal mix of models for your organization, you’re definitely not alone. Many analysts conclude very few have this figured out top to bottom, and the reality is companies operating with clear, defined strategies is increasing, but more slowly than you may think.”
The time to act and define your cloud strategy is now. Enterprises that take time to educate themselves on the advantages and disadvantages, assess their appetite for control and willingness to shoulder heavy lifting, and select the service models that best fit their needs will maximize their cloud investment into the future.