Networks have traditionally supported applications by providing a transport service. Over the years this transport service was optimized through quality of service mechanisms in an effort to prioritize one set of applications over another. After more than ten years of web development and investment, the type and number of applications have increased dramatically, driving new application support requirements within networks. A new class of network service called Application Intelligence has emerged to address a series of problems and provide IT leaders with tools to optimize application performance while increasing user experience.

One example of Application Intelligence can be found in routers today via Cisco´s Network-Based Application Recognition (NBAR). To expand NBAR´s functionality reach throughout a corporate network and increase performance to the multi-gigabit level, Cisco introduced its Programmable Intelligent Services Accelerator (PISA) into the Supervisor Engine 32 for the Catalyst 6500 series of switches on April 30th. There are appliance-based Application Intelligence solutions as well from various vendors such as Packeteer, F5, BlueCoat, RiverBed, et al.

Application Intelligence is being deployed at the network access and the DMZ, creating an Intelligent Edge and Intelligent DMZ. The Intelligent Edge surrounds a network at the wiring closet level, classifying applications, assigning priorities, defending against day zero exploits and providing IT executives with application and network behavior information. The Intelligent DMZ provides protection and defense for web servers against attackers while alerting operations to suspicious traffic flows. Application Intelligence provides insight into internet ingress and egress traffic flows and policing of application types, eliminating unsupported applications before they traverse over the internet optimizing internet bandwidth for mission critical applications and appropriate use. The Intelligent Edge and DMZ prioritize mission critical applications preserving IT investments and optimizing business process while delivering corporate material benefit.

Application Intelligence has become available at the right time, as new application delivery models have emerged, representing unknown and unforeseen traffic patterns and network loads. Structured IT applications are converging upon Web services, which make it difficult to distinguish mission critical from recreational applications. Unstructured applications based upon Enterprise 2.0 technologies like RSS, AJAX, mashups, wikis and SOA network behavior is unknown. Thin client and back-end-based application delivery is yet another network unknown. Add convergence or unified communication and TelePresence real-time network requirements and Application Intelligence provides the application view to manage and navigate through this unchartered industry phase of application delivery.

In the old days IT executives would throw bandwidth plus core switch and routing processing at the problem of poor application performance. Others would add bandwidth plus spend large amounts of time implementing QoS mechanisms in both the LAN and WAN. Application Intelligence offers a different approach to campus network design by automating application performance management. Where there are multiple different applications flowing across the network, they can all be logical, meaning that a network is made up of ?¬¢‚Äö?ᬮ??¨n?¬¢‚Äö?ᬮ¬¨?? logical networks with Application Intelligence managing these logical entities. Application Intelligence is managing the connection between applications be it a strong or weak connection, or no connection at all. Application Intelligence allows the application to view the network as though it is matching the network´s capabilities with its unique requirements. In short Application Intelligence allows every application to obtain its fair share of resources, bandwidth, QoS, and Service Level Agreement (SLA) in the presence of all other applications.

As Application Intelligence´s value is greatest when deployed pervasively, a five stage Application Intelligence implementation process is offered in the white paper “Application Intelligence A New Network Service“.

Based upon preliminary empirical data, i.e., customer experience, the cost to equip an average campus network with Application Intelligence is 14% of capital spend associated with a total refresh of network switches. Application Intelligence will be bundled with new switch procurements eliminating the need for separate Application Intelligence budget development. With Application Intelligence spend at approximately 14% of switch acquisition cost, IT leaders gain the benefits of application classification, visibility, policing and end-to-end QoS while business leaders gain increased security, regulatory compliance and a competitive advantage through improved business process performance and optimization.

Posted 21 May 2007 at 12:47 pm under Lippis Report. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.