The Lippis Report

As business and IT leaders think through their next generation branch IT infrastructure many are reviewing their IT service delivery model, thanks to the third phase of integrated networks and applications, which we discussed in Lippis Report 111. The number of branch office locations is increasing and will accelerate thanks to the financial market crisis which has transformed Goldman Sachs and Morgan Stanley into bank holding companies, allowing them to take deposits and expand their presence throughout the globe. But independent of the financial market dynamics business and IT leaders have been re-distributing human and capital assets away from headquarter facilities toward branch offices for some time now and the network-embedded application approach allows IT leaders to manage this shift in business operations.

Strategies that leverage data center consolidation efforts, which minimize branch office IT footprints and TCO are favorable now more than ever. Server centralization, virtualization and network consolidation are key aspects of next generation branch office IT infrastructure. Application deployment, integration and performance are key management focus areas as is branch survivability, and the industry is responding to IT leader requirements by offering a new network-based application service delivery model.

Writing and residing business applications in network gear offers a new service delivery model where dedicated compute and storage capacity reside within a branch office router. This integration is a new application delivery platform offering several value points: 1) through an API branch applications gain network dynamic insights so that application performance is optimized; 2) networks, server and application consolidation in the branch; 3) both vertical and horizontal specific applications can be delivered via this platform; and 4) a new application delivery ecosystem is emerging, easing branch application deployment.

The integrated application platform enables independent software vendors (ISVs) to offer branch specific applications for targeted vertical industries such as healthcare, financial services, retail, transportation, et al. In addition ISVs are porting their software to this new platform to address horizontal or cross-industry branch office requirements.

For a horizontal application example consider Sagem-Interstar, a global leader in fax server solutions for IP networks. Sagem observed that legacy fax machine deployment is both expensive (they required dedicated analog circuits) and do not preserve privacy. Sagem realized that it could deploy its XMediusFAX Fax over IP (FoIP) technology on Cisco’s AXP to virtualize fax functions into the Cisco ISR VoIP gateway. By doing so any desktop application could send/receive faxes; plus fax services could now be integrated into e-mail and mobile devices. Not only is cost reduced by running XMediusFAX on Cisco’s AXP, but AXP improved resource utilization and provided audit trails for compliance requirements.

This solution came together for H I G Capital. H I G Capital is a leading global private investment firm with over $7.5 billion of capital under management. H I G created value and optimized its remote office operations through Cisco’s AXP business platform, which tightly links applications and networks. In the podcast linked below Luis Suarez, Director of IT at H I G Capital explains how H I G used AXP along with Sagem-Interstar’s XMediusFAX application to displace both an old world analog fax network plus web-based fax services, while integrating fax services into its unified communications environment extending fax services to all H I G end-points and in the process speeding up business process and deal flow.

H I G was able to remove all fax machines and their associated maintenance contracts, and improve service for company professionals without the need to add new equipment to their branch offices. The FoIP service was delivered by IPComIT, a Sagem and Cisco partner.

In addition to ISVs and application providers porting their applications to the branch application platform, in-house IT development departments such as those in the financial services industry are writing custom applications, which increase application performance or add features to management tools. Integrated application branch platforms offer APIs in which developers can write their own applications to add value to their branch office IT infrastructure.

The Value Proposition of Business Applications Written to and Residing in Network Gear

The value proposition for the network-embedded applications is specific to branch office IT infrastructure as it leverages data center consolidation while increasing IT services at lower TCO. The network- embedded application approach increases branch office IT infrastructure attributes thanks to a new application delivery model that lowers branch office IT footprint by consolidating networks, servers and applications onto one platform reducing the IT operational cost.

Lower TCO: Depending on the number of branch applications, multiple servers may be needed, but their total number will reduce in this model. The highest cost components of branch office IT TCO (operations and facilities) are reduced with a common management model for networks, servers and applications. Maintenance is reduced by the displacement of devices such as fax machines, voice recorders and some servers while WAN performance is optimized by the interaction between network code and applications plus WAN application acceleration.

The following four findings from a recent study compared application development of in-branch offices appliances/standalone servers versus the network-embedded application approach.

• Quicker Application Development Time: Application development cost is significantly reduced when developers utilize an integrated application branch platform versus an in-branch server. Because the network is being utilized as a platform, there are savings in terms of time and development costs thanks to integrated services plus libraries and APIs, to which developers have access. The application is able to work directly with the network and leverage its APIs. Time required for development of new features in a network-embedded application scenario accounts for approximately one-third of development costs. The integrated application approach reduces this development time by approximately half.
• Lower Integration Cost: Integration cost represents time involved in configuring and integrating the application server, security device, branch office router, etc. It’s the cost of bringing together a heterogeneous environment to allow transparent operation to achieve business goals. In the network-embedded application scenario integration consumes approximately one-fourth of the effort to own and operate an integrated application branch platform versus appliance or separate branch office server.
• Lower Maintenance and Energy Consumption: Maintenance includes the time and costs related to implementing patches, upgrades and updates which grows in time, cost and complexity as the number of devices in the branch increase. The network-embedded application approach reduced maintenance cost by over 50%. Power and cooling cost are also lowered by some 50% thanks to the integration of networkings, computing and storage in one package.
• 4) Faster Business: As servers are resident in the network and applications have visibility into network dynamics via APIs, faster deployments are achieved for new applications. New applications can be deployed without the procurement, configuration and operations of a new server, reducing the time required to launch a new branch office service. As more applications are layered on top of the integrated branch server platform, the time and cost associated with maintenance is reduced as well. This quicker pace of application deployment and efficient maintenance allows an organization to respond quicker to market dynamics and/or customer demands/concerns.

In addition to the above four advantages of network-embedded applications versus separated, survivability is increased as applications are able to respond quicker to changes in network status.

Survivability: There are specific network attributes gained through new design options afforded by the network-embedded application approach. In particular applications become network aware. For example, if the LAN link is performing sub-optimally, the application platform can sense this and dynamically alter some of the router settings based upon business load so that sensitive applications stay within performance tolerances. In the wide area, a failure of a primary or secondary link can be mitigated by configuring the two in a load-balancing manner that leverages performance routing and increases application survivability and performance.

Network Smart Proxy: Many branch applications are hosted in a data center and accessed over the branch office network. Some applications provide proxies in the branch office to increase survivability. There is a continual process of checking status between the network-embedded application platform and its clients plus the central application to detect a network failure and if one occurs to assume the role of proxy, assuring business continuity. Data center consolidation allows IT leaders to centralize complexity and deploy application proxies into branch office locations. That is, the application resides in the data center and the proxy could be running in the Cisco AXP; for example, in the event of a link failure or disaster the application platform could keep the proxy running until such time as access is available to the data center, transferring control back to the central entity.

Network and Application Awareness: Not only are the applications aware of the network, but the network is aware of the applications. Seems intuitive, but this awareness between networks and applications allow both to behave respectfully of each other so that performance and survivability are improved. This awareness improves disaster recovery and business continuity as the network and applications have a view into and some control over each other allowing quicker response to both man-made and natural disasters. In addition most IT organizations are split into application and network groups. With awareness between networks and applications, IT is able to view the service delivery, acquisition and performance optimization more holistically.

A tight linkage between the application environment and wide area network offer additional performance and disaster recovery attributes. For example, application performance over the wide area is improved as it has a view into router status and network dynamics, which it can manipulate based upon business rules. Applications can avoid network-based delays such as shifting application traffic from a heavy traffic WAN link to one less occupied. Not only can applications shift their transport based upon performance, but also failed links are avoided. As mentioned above application driven router reconfiguration allows application to modify network behavior to maintain performance. IT leaders may centralize most application support and create a very thin or thick branch application footprint, depending upon their business needs. The network- embedded application approach affords optimization of both models yielding great control over IT complexity.

One of the most important attributes afforded by the integrated application approach to branch office IT infrastructure is the speed of response to market conditions and the dynamics it offers to IT leaders. As executives seek to capitalize on globalization opportunities, manage IT complexity, become greener and deploy Web 2.0-based collaboration, the network-embedded application approach is an enabler of these important initiatives. As business and IT leaders re-distribute human and capital assets away from headquarter facilities toward branch offices the integrated application approach allows IT leaders to manage this shift in business operations.