on 29-Apr-2011 02:59
It’s not just cloud computing and virtualization that introduce volatility into the data center.
But it isn’t just the volatility of cloud computing and virtualization that makes traditional data center architectures brittle and more apt to fail. It’s the constant barrage of users, devices, and locations against a static data center configuration that makes a traditional architecture fragile and inefficient.
Pressures are mounting both from within and without on data center infrastructure to assure availability, security and high-performance of applications to every user, regardless of location or device type. With the rapidly changing landscape of devices and smartphones and locations of users, this means an ever changing array of policies governing access and assuring availability.
The processes by which these are enforced is not sustainable in the face of such growth. The burden on operations, network and security teams can only grow based on such a static model. What’s needed to address the dynamism of the user and application environment is a dynamic infrastructure; a dynamic services model that adapts to the increasingly complex set of variables that can impact the way in which infrastructure should treat each and every individual application request – and response. It is this characteristic, this agility in infrastructure, that is critical to the implementation of a dynamic data center and an agile operational posture. Without the ability to adapt in an intelligent and programmable fashion, operations and infrastructure cannot hope to scale along with the growing demands on the network, application and storage infrastructure.
It is important to stress that the dynamic services model is an ecosystem approach; it is not a single vendor solution or a point product. The programmatic and procedural resources provided to integrate, coordinate, and collaborate with other ecosystem elements are defining characteristics of F5’s dynamic control plane. This is evident in the illustrated solutions with VMware and Gomez, but these are only a few examples of the F5 dynamic control plane solving real-world problems today. F5 maintains formal relationships with leading technology providers including Dell, HP, IBM, ExtraHop, Infoblox, NetApp, CA, Symantec, webMethods, Secure Computing, RSA, WhiteHat Security, Splunk, TrendMicro, ByteMobile, Microsoft, and many others. These relationships include tested and documented integration with F5’s solutions and, as such, can be thought of as an extension of the dynamic control plane architecture.
-- Ken Salchow, “Unleashing the True Potential of On-Demand IT”
Within F5 we refer to this strategic point of control as the dynamic control plane; a platform that is adaptable, programmable and intelligent with regard to both its run-time and configuration-time operations. The full-proxy nature of F5’s underlying application delivery platform, TMOS, provides an interconnected and contextually-aware environment in which requests and responses can be collaboratively intercepted, inspected and if necessary, modified to assure the highest levels of availability, security and performance. By providing a common high-speed interconnect that shares context, F5 BIG-IP solutions are all capable of understanding not only the context of each individual request and response, but the business and operational requirements placed upon the data in a way that allows the platform to make real-time decisions regarding policy enforcement.
From a deployment perspective, the dynamic control plane enables an agile operational posture; one that integrates via a standards-based, service-enabled API to provide the means by which BIG-IP can be integrated and collaborate with other data center management platforms to provide automated provisioning, context-aware monitoring, and infrastructure as a service. By enabling the platform with a common set of remote management interfaces, BIG-IP can be managed, monitored and informed through collaborative technologies that increase its abilities to make informed decisions regarding ingress and egress traffic such that the appropriate policies are enforced at the appropriate time on the appropriate end-user, device and location. Combining end-user, network, and application awareness means BIG-IP is enabled with the data necessary to adapt in real-time to conditions that exist now rather than conditions as they were five, ten or thirty minutes in the past. While historical trending is helpful in setting appropriate policies the ability to react quickly means unanticipated variables can be accounted for more rapidly, which means less time in which possible outages or breaches may occur.
The results of such collaboration can be seen in joint solutions such as:
All of these solutions (and others) take advantage of F5’s dynamic control plane both for automating the processes necessary to achieve the level of dynamism required as part of the solution to the challenge and for implementing the decision-making processes required at run-time to address the dynamism that drives the need for those processes to exist.
We’re trying to get to the point where IT has the ability to react to conditions as they change in a way that enhances the performance, availability and security of the data center as a whole. It’s about programmability and processability, about being able to specify policies to address “what-if” scenarios and then trusting that those policies will be enforced in the event they come to fruition. It’s about making infrastructure as agile as the conditions under which they must constantly deliver applications, and doing so as efficiently as possible.
A dynamic services model enables operations to assume a more agile posture regarding deployment and delivery of applications. F5’s dynamic control plane makes it possible to do so efficiently, intelligently and collaboratively.