Cloud Computing: Far More than Infrastructure
Cloud computing sets a new standard for infrastructure expectations. In the past, gaining access to computing infrastructure commonly took weeks, months, or even years, depending upon the vagaries of corporate budgeting, project priority, and staff availability. With cloud computing, infrastructure expectations are now measured in minutes: the big three providers (AWS, Azure, and Google Cloud) make it possible to go from initial account setup to running virtual servers in less than 10 minutes.
Developers and applications groups rightly celebrate this innovation. They’ve used this rapid availability to accelerate application development and rollout. They’ve used it to improve quality, because infrastructure is no longer in short supply; infrastructure rationing typical of the old model starved groups like QA and operations. They’ve even used it to pursue innovation, because fast infrastructure instantiation and teardown makes it easy to experiment and try new initiatives -- and quit them quickly if they don’t pan out.
However, the most farseeing of developers and application groups recognize that cloud computing is far more than fast infrastructure. They understand that infrastructure is only the foundation of cloud computing, and just as much value resides further up the cloud computing stack, where other services accelerate the very process of software development -- which supercharges the application lifecycle.
We can see the evolution of cloud computing thinking in this blog post by Drew Firment of Capital One. In it he points out that cloud computing forces an IT revolution in computing. But he also points out it enables a rethinking of how a company delivers value to its customers -- and that customers don’t really care about a company’s infrastructure. The big shiny datacenter that the company runs? It adds no value to the customer experience.
Firment goes on to make an even more important point that many IT organizations -- and the employees that staff them -- fail to recognize: that customer indifference to the company’s computing extends to everything other than the delivered value experienced by the customer.
As he notes regarding the fancy DevOps toolchain that many IT organizations are developing to make their application pipeline as fast as the underlying infrastructure:
- The explosion of unique combinations of CI/CD tools throughout an organization are both unnecessary and counterproductive. The majority of enterprise DevOps continuous delivery pipelines are science projects that work locally, but are detrimental to the well-being of the overall system you’re working within.
- Similar to the commoditization of infrastructure into a compute grid by AWS, a similar story is playing out with the DevOps pipeline. With the announcement of CodeBuild as a new addition to the existing suite of AWS Developer Tools, there are less and less reasons to roll your own pipeline.
In other words, IT organizations should examine their entire value chain -- how they go from an idea for an application to the delivery of functionality to a user -- and ruthlessly adopt whatever is the fastest, cheapest, most commoditized way of delivering value. Any part of the IT value chain that doesn’t differentiate the organization’s parent company in the marketplace is ripe for replacement by services offered by a low-cost provider, which is nearly always a scale cloud provider.
This approach allows the IT organization to focus its efforts and budget on the part of the value chain that delivers unique functionality to the customer.
Adopting this approach requires IT organizations to understand that cloud computing is far more than outsourced cheap and fast infrastructure. Cloud computing is a rich set of computing services that can be assembled to construct an application quickly. More important, every service an application group leverages is one more part of an application that the group isn’t responsible for; every responsibility off-loading allows the group to focus more of its efforts on customer value.
That’s why the certifications that Simplilearn offers address much more than what kind of virtual machines a cloud provider offers. Certificates address storage options, notifications and email services, application frameworks, and operations monitoring -- the full gamut of a cloud provider’s services.
It’s often difficult for IT organizations that have spent so much time mired in obtaining and managing infrastructure to recognize that cloud computing upends all the traditional assumptions associated with delivering applications. Firment’s post is a strong reminder that IT organizations need to redirect their assumptions -- from believing their job is managing computing resources to recognizing their charter is delivering customer value.
Moreover, IT organizations must realize that this requires evaluation of the entire value chain and focusing on those portions that only the organization can deliver. That means handing off responsibility for as much of the computing stack and possible and spending inevitably-constrained staff resources on company-specific functionality.
About the On-Demand Webinar
About the Webinar