This article originally appeared in the April 2012 issue of INTERNET TELEPHONY magazine.
For the past few columns, I’ve been focusing on various topics and technologies surrounding cloud bridges, gateways, and brokers: what they are; why you should care; what they can do; where you should use them; etc. In addition to cost savings, mobility is one of the largest drivers behind enterprise IT moving to the cloud, and an extension of that mobility is the ability to chain a virtual cloud data center off of a standard physical one.
Analogous to the shift away from physical resources within the data center, cloud bridges and gateways allow enterprise to move away from physical connectivity, such as dedicated links and dark fiber between data centers, and virtualize the last mile of the cloud: connectivity. But the end-goal isn’t just connectivity; it’s the ability to use that connectivity as part of a single data center resource.
Virtual cloud connectivity devices – in particular, cloud gateways – allow us to connect not only the resources in the cloud to the on-premises data center but also to integrate the remote services of the cloud. This extended connectivity model enables us to create a transparent and fluid hybrid cloud model for users and applications: the end-goal of any hybrid cloud deployment. We want to be able to move any part of any application into the cloud, transparently re-direct users to that cloud-based application as needed, and manage a single access, availability, and security policies between the on- and off-premises data center resources. This model – a single application namespace – may seem like a stretch goal, but it’s not; it’s completely achievable today using cloud gateways, and without much extra work for the enterprise.
Thankfully all the devices and components required to support a single application namespace exist today: presentation systems that track where the application currently lives (in the cloud or at the on-premises data center) but represent the true location as a single virtualized location to the customer (think DNS); customer portal systems that offer up multiple options for delivering enterprise applications to mobile users, such as single sign-on VPN devices that always live at the same hostname but dynamically distribute users to the back-end applications in the cloud depending on load, location, availability, device, etc.; infrastructure components, such as cloud gateways, that manage dynamic routing, verify that cloud-based applications are available and are the best option for the end users; the list goes on and on.
The technology is readily available today, but most of these technologies are available as single-point solutions spread across multiple vendors offering many different software and hardware options. This disaggregated model makes managing a single application namespace across the cloud nearly impossible.
One of the reasons is a diversified marketplace. Enterprise IT is over-inundated on a daily basis with “cloud-ready” infrastructure options, many which require a complete overhaul of core networking in the on-premises data center and/or have limited support between multiple off-premises cloud providers. At the same time there is a rush for cloud providers to support enterprise customers by offering unique and differentiating value-add solutions – which means opening up new services that support the enterprise in different ways than the now-standard cloud model of spinning up services with a credit card then walking away. So enterprise IT is forced to choose a particular technology that will both serve their feature needs and support the dynamic environment of multiple cloud providers. Or they go the other way and first choose a cloud provider and then drill down into what, if any, solutions they offer in the way of cloud connectivity.
The true solution to a single application namespace is a cloud gateway that is integrated into both the on-premises data center as well as the off-premises cloud provider, can support the dynamicity of bouncing applications and users between multiple data centers and different cloud providers, and has a goal of delivering a single application namespace throughout the life of the cloud-based application.
The critical error most IT departments and cloud providers are falling into is focusing on network connectivity. They should be focused on managing the end-to-end application deliver solution from the cloud. The cloud was built as a new model to deliver applications to users, cheaper and faster. In order to take full advantage of applications in the cloud, we need to start moving, en masse, to cloud gateways that provide a single application namespace – transparently delivering any app to any user from any location without sacrificing IT control and management.Edited by Jennifer Russell
