Private clouds require a Capex investment ahead of demand, which means companies actually need to make a larger investment than does a company with the traditional IT deployment model for compute resources. Thus, the initial cost of deploying infrastructure is not where the private cloud model yields its benefits; the operational efficiencies of deploying a cloud and consuming its resources over the entire lifecycle are where
Enterprises must take advantage of the difference between a private cloud and virtualized infrastructure to capitalize on benefits.
There are at least four areas where a private cloud can enhance IT efficiency. They include:
1. Purchasing: If you procure multiple instances of the same server stockkeeping unit (SKU) in a single transaction, the vendor will offer volume discounts. Also, cloud compute servers are usually specified to take advantage of mainstream pricing rather than pushing performance limits. High-performance applications will use multiple moderate performance compute units rather than a single high-performance unit.
2. Installation: The inherent modularity of cloud infrastructure makes the installation of new resources more efficient than in traditional IT. A module may be of any size, but "rack scale" is very convenient for most enterprise data centers. At a minimum, a module contains compute servers and the network to connect them. It is possible to outsource the assembly of the rack, which further streamlines both purchasing and installation, and repair can also occur at the module level.
3. Resource sharing: Many IT workloads occur on a schedule; different workloads occur on different schedules. Typically, resources are dedicated to each workload, which leads to poor utilization because resources sit idle a large part of the time. Properly architected cloud applications use the same resources, which greatly improves utilization. Other private cloud scenarios include temporary or seasonal workloads, in which sharing is not as predictable, but, currently, it represents a huge opportunity to improve efficiency.
4. Management: A private cloud finally delivers on the automation that virtualization promised. With a clearly defined service catalog -- that includes compute, storage, database, etc. --developers not only acquire the resources they need to build and test an application, they also know at design time what the production environment will be. A quick visit to a cloud portal delivers those resources with zero involvement from IT personnel. Once the application is in production, automation enables seamless scaling in response to changes in demand.
So where are enterprises missing out? The way enterprises currently approach the private cloud model is reminiscent of the definition of insanity attributed to Albert Einstein: doing the same thing over and over again and expecting different results. If an enterprise deploys a private cloud and uses it the same way it uses its existing virtualized infrastructure, that enterprise should not expect any significant improvement in efficiency.
The same is true of cloud washing an existing data center, as nothing has changed and the end result is quite literally "doing the same thing." Instead, enterprises must take advantage of the difference between a private cloud and virtualized infrastructure to capitalize on benefits and achieve different results.
The single biggest change that has to happen to reap the benefits of a private cloud is that applications must target the infrastructure, instead of the infrastructure conforming to the applications. Enterprise application architects need to make this change for internally created business applications, and ISVs need to change how off-the-shelf software is deployed.
This is a strategic shift in how business applications are installed and requires a team effort. Developers consume the cloud. And executives must be patient while their organization makes the evolution to achieve truly different results. Without support from the top, politics will disrupt the creative processes required to break free of the old paradigm.
About the author:
Mark Eisenberg has been doing mobile app development since 2005. He joined the nascent Windows Azure sales team at Microsoft despite being an early cloud skeptic. Now, after embracing the cloud and its technological potential, he combines his cloud and mobile expertise with his technological background to help clients realize real value from their technology investments. Mark is also a seasoned business development professional with over 20 years of experience. He started his career in software development and has since maintained his technology edge, most recently adding cross-platform mobile development skills. His sales career began when he joined Intel's channel and has included positions at other communications-focused firms prior to Microsoft.
This was first published in January 2014