There are plenty of good reasons to push the enterprise infrastructure model onto the cloud. It is cheaper, more flexible, and is becoming increasingly more secure and reliable as both hardware and software architectures evolve.
But there are plenty of good reasons to maintain a certain amount of on-premises infrastructure as well, and as time goes by, the deployment and management of that infrastructure will become vastly less complicated and expensive than what you find in today’s bloated data center.
Enterprise Storage Forum’s Drew Robb recently posted no less than 12 good reasons to keep some storage on-premises, depending, of course, on your performance needs, data volumes and other characteristics. Frequently accessed data, for example, is much better off when housed close to the knowledge workers who require it rather than in some far-off cloud facility. Highly critical data may also need to be kept close, not because the cloud in general is any less reliable, but providers themselves may cease operations at a moment’s notice, forcing clients to overload IP connections trying to get their data out. As well, connectivity costs in the cloud can be unpredictable at times, leading to less certainty in overall costs.
Part of this new cost-benefit paradigm is due to the new ways in which enterprise infrastructure is being designed in response to the cloud juggernaut. Cisco, for instance, claims its on-premises storage portfolio can produce a 50 percent better TCO than the cloud, driven primarily by its new S-Series Unified Computing System (UCS) portfolio. With a 4 RU modular system now capable of scaling out to 600 TB, Cisco says it can cut capex by 34 percent, lessen maintenance costs by 80 percent, cut cabling by 70 percent, and consume nearly 60 percent less power than equivalent capacity on a traditional array. Added up, says the company’s Todd Brannon in a recent interview with IT Trends & Analysis, you can store 420 TB over three years for about $250,000 vs. $550,000 for Amazon S3.
HPE makes pretty much the same argument for its new line of “composable infrastructure” products. The company’s Gary Thome says that by offering cloud-like capabilities using modular, interchangeable hardware components, the enterprise can dramatically lower its own IT costs by expanding its support for highly flexible workloads and broad scalability. Under the composable model, infrastructure is converted into fluid resource pools that support near-instant deployment to suit a wide variety of highly specialized application requirements. At the same time, advances in software-defined architectures and artificial intelligence will strip much of the complexity away from IT management, allowing knowledge workers to employ API-driven self-service management to compile resources as needed.
Even Amazon is starting to rethink the all-cloud strategy that has driven its system development over the years. The company undoubtedly still thinks it provides the optimum environment for the bulk of enterprise workloads, but with releases like the recent Amazon Linux AMI (Amazon Machine Image), it also recognizes that some infrastructure will remain in the enterprise, at least for the time being. The system, according to E-Commerce Times, allows users to deploy the Linux AMI on internal resources, essentially providing tighter integration between local infrastructure and Amazon data centers. In this way, even high-performance applications can maintain a stable and secure link across hybrid deployments which, according to Amazon, would be most useful when developing and testing new apps and workloads at home before deploying them in the cloud.
Whether on-cloud or on-premises, however, the final call will be made by the enterprise, which will now have to take into consideration its workload and application requirements, cost factors, performance goals, and a host of other elements for each and every deployment.
This will be one of the true ironies of the cloud era: Even as management and maintenance becomes simplified, deployment and configuration responsibilities will mount.
Arthur Cole writes about infrastructure for IT Business Edge. Cole has been covering the high-tech media and computing industries for more than 20 years, having served as editor of TV Technology, Video Technology News, Internet News and Multimedia Weekly. His contributions have appeared in Communications Today and Enterprise Networking Planet and as web content for numerous high-tech clients like TwinStrata and Carpathia. Follow Art on Twitter @acole602.