Convergence seems to be the new black in the enterprise.
That may seem like an odd statement at first, considering convergence has been trending for several years now. But unlike fashion, which comes and goes by the season, technology has a way of playing out over several years. And while most people may have a clear idea of what convergence means in the enterprise, the fact is that it has taken on a number of surprisingly new forms of late.
Traditional convergence, of course, kicked off with the rise of server virtualization. For the first time, the enterprise could envision not only higher utilization rates and streamlined server architectures, but the end of the much maligned silo architecture that hampered the flow of data across divisions or even individual business units. As The Taneja Group’s Mike Matchett notes, traditional domains have been abstracted into fully realized compute/storage/networking entities, which can be scaled, mixed and matched largely on the fly to accommodate rapidly changing data environments. The next step is “hyper-convergence,” in which fully modularized infrastructure is stacked, pooled and scaled as needs dictate.
This kind of convergence generally revolves around traditional enterprise systems. But the term is increasingly being used to describe the integration of seemingly disparate infrastructure into a more cohesive whole. Frost & Sullivan, for example, recently issued a report detailing the convergence of network, wireline and wireless access and applications. In this world, multiple traffic types are combined on a single network framework, which can then be optimized through bandwidth utilization and allocation techniques to improve productivity and simplify the implementation of new technologies as they become available.
In many ways, according to CIO.com’s Allen Bernard, modern convergence is a conglomeration of previous streamlining efforts, such as grid or utility computing, on-demand or shared services, and even the cloud. So far, at least, this complexity hasn’t scared off small and medium-sized enterprises, and in fact may be the key to helping them compete with larger rivals. Of course, technology developments have a way of solving old problems while creating new ones – anyone who saw what virtualization did to storage infrastructure can attest to that. But for the moment, at least, flattened infrastructure seems both easier to manage and less expensive to build and operate.
For the major platform vendors, convergence is a lifeline in a time of rapidly dwindling hardware sales. HP recently upgraded its Converged Infrastructure portfolio with a new BladeSystem enclosure and converged storage and networking solutions, while Dell has teamed up with VMware to develop a converged solution designed to augment desktop virtualization infrastructure. Both companies seem to have finally recognized that while volume shipments are highly profitable, converged infrastructure is a more attractive proposition from the enterprise perspective. And as the saying goes, “the customer is always right.”
It would seem, then, that convergence in all its forms is likely to remain a guiding force in data center development. As the enterprise becomes accustomed to the low cost and flexibility of the cloud, internal infrastructure will have to provide an equally appealing cost-benefit proposition in order to remain viable.
That means it will have to slim down and shape up in very short order.