Let’s face it right now the cloud is pretty immature. The level of automation and management of these environments are analogous to the early assembly lines, but it won’t be this way long. This is not the industrial revolution and it moves at a wicked fast pace. Before we know it the next generation of cloud computing will be upon us and it will be very different than the IaaS/PaaS/SaaS offerings we know today.
For one, it will be intelligent. That is, the cloud will be content aware and it’s network connections will act like mycelia hyphae and what one hyphae learns will become available to the entire cloud. Whereas the current cloud is focused on scalability and elasticity, the next instance of the cloud will focus on redundancy, resiliency and collaboration. The discussion regarding public, private or hybrid will become moot as the cloud simply becomes a system of nodes with some nodes participating fully while some don’t participate at all. Nodes will contribute to the cloud on a controlled basis. Some will host their own nodes while others will pay service providers to host their nodes for them.
However, the bigger issue is not when will this occur, but why must this occur? This must occur because we are learning that no matter how much the cost of compute resources comes down, it will never be enough low enough to be cost-effective to host the Zettabytes we’re interested in. The cloud today is teaching us a valuable lesson; content is king! Once we squeeze all the inefficiency and underutilization out of our data centers there will be little cost savings left to derive from our own cloud infrastructure, but that won’t stop the machine once it’s started. Just like any other successful ecosystem, once started, it eats foundation and then starts feeding externally to survive. This pattern is how small companies becomes large corporations. This pattern is how small republics become big government. The cloud as an ecosystem is no different and it feeds on content. When it consumes all the content we can provide it with about our own organization, it will start to feed on external content. We are starting to see this occurring already under the guise of “Big Data”.
The current focus on what is cloud computing is but a mere distraction fostered by a market that is organically moving toward the culmination of the intelligent cloud. This, however, doesn’t undermine the effort underway as it is a critical component of reaching the intelligent cloud outcome. That is, the consolidation of silo compute stacks onto converged infrastructure is a critical first step toward the node architecture of the intelligent cloud. However, the lack of discussion and focus on application rationalization will have profound effect of limiting forward progression. Moreover, the limited tools for inventorying and understanding the dependencies between application components forces the application rationalization process to be heavily based on human knowledge engineering.
Until the tools market for application rationalization matures, it is imperative that organizations get serious about building their Configuration Management Databases (CMDB) and following IT Service Management processes. Failure to comply with these imperatives will significantly limit the upside advantages that cloud computing can provide to the business. Sure, executives will be thrilled with the immediate cost reductions, but when was the last time anyone remembers their CEO saying two years later, “don’t worry Bill, you saved us $2 million two years ago, you’re still golden in my book!” The immediate cost savings from infrastructure consolidation, SaaS outsourcing and Big Data analytics will be short-lived and the CEO will be looking for when they can finally start to sunset some of those proprietary application stacks and move their applications to their costly cloud infrastructure, only without the tools and without the ITSM foundations, the answer is going to be that it will require big up-front spend to gain efficiencies and further costs savings in the future.