It all sounds so simple in theory: A highly dynamic cloud that allows the enterprise to shift loads, applications and virtual resources across the globe to take advantage of peak performance, low operating costs and a host of other desirables.
The trouble is, good intentions tend to give way to cold, hard reality. And the simple fact is that while a fully federated, fully automated cloud is still within the realm of possibility, getting there won’t be as easy as once thought.
A key piece of the puzzle is the application delivery controller (ADC), which performs a number of crucial functions in the dynamic architecture, such as load balancing, multiplexing and security. The ultimate goal is to foster an environment in which data can seek out the most efficient support infrastructure on its own, making the type and location of hardware resources largely irrelevant to both the user and even the IT manager.
Naturally, development along these lines is highly sophisticated, so much so that even top platform providers are leaving it to the specialists. Cisco announced this week that it is shelving development of its Application Control Engine (ACE) for Catalyst switches and routers, citing the need to “align investment based on growth opportunities,” which many interpreted as a nod to the fact that smaller firms like F5 are already running away with the ADC market.
Indeed, the announcement led to a flurry of moves to capitalize on Cisco’s exit. A10 Networks, for example, is offering up to $24,000 in installation and migration services for ACE users who switch to the AX Series ADC. F5, Citrix and others are also said to be shoring up their channel partnerships and promotional strategies to tap into the largely all-Cisco network environments that made up the bulk of ACE deployments. Meanwhile, Brocade seems committed to its ADC platform, the ADX Series, which the company considers a crucial component to its development of the software-defined network.
At the same time, a number of startups are poised to take advantage of the need to optimize application delivery in the cloud. Nixu Software, for example, just released the Nixu Cloud IP Suite that employs a technique called “release parameter provisioning” that automates the application deployment process, cutting it down from several weeks to mere seconds. The system features tools like IP address management, open APIs and workflow/infrastructure integration services that streamline application delivery processes without altering existing production environments, even in multi-vendor/multi-tenant deployments.
All of this is part of the continual evolution of the ADC from a simple load balancer to an all-purpose application optimization solution, according to Array Network’s Paul Andersen. Along the way, it has transformed from an appliance-based solution to commodity hardware and finally a cloud service. At its heart, however, an application delivery solution must provide three core competencies — app delivery, access security and wide-area optimization — no matter what form it takes or how it is deployed.
Application delivery is the key to turning the cloud from primarily a massive storage architecture to a working, highly scalable version of today’s data center environment. We’re still at the very beginning of what is likely to be a lengthy development process, but in the end it will produce the kind of broad-based application and data flexibility that cloud boosters have been promising for the past three years or so.
And whoever devises the breakthrough platform for app delivery is likely to become a major player in the future development of information technology.