Lippis Report Issue 105: What I Learned At Interop

The Interop expo and conference used to be an industry event where you got an industry snapshot and projected major trends for the year to come. But the organizers diluted the event by adding the software2008 and CSI SX security conferences; plus they offered conferences within a conference such as their CIO Boot Camp, Energy Camp, the Unconference, et al while piling on IT topics for the main Interop conference. In all there were over 250 conference sessions on nearly every IT topic one could image or create spread out over five days. The exhibition floor attendance was about the same as last year +/- 10% at approx 20,000 attendees. With so much noise and very little signal at Interop, you had to ask what was the value of attending? For me it's the one-on-one meetings. This is what I walked away with from Interop.

1) Ports Are Dead: In the networking business port counting has been the organizing principal used to calculate market share and assess which supplier was gaining and losing share. Industry analysts would count every different type of port –10MbE, 100MbE, 1GbE, 10GbE, switched ports, shared ports, layer 2 ports, layer 3 ports. You named it, they counted it. But as the IT industry accelerates its pace of virtualization adoption where a single large 10GbE port supports numerous logical ports and flows to a single blade running multiple virtual machines, the definition and meaning of ports will change to virtual ports.

2) Data Center Virtualization Shifts Networking Back Into Computing: Over the past two and a half decades networking has been off-loaded from computing. Many networking functions such as switching, routing, load balancing, caching etc were once done in computing systems as they ran applications. The networking industry grew out of the segmentation or distribution of networking tasks to specialized devices such as switches, routers, firewall, load balancers, etc. With the increased use of VM and other virtualization approaches, networking has been cycling back to computing. This is most evident as tasks flow between virtual machines through a hypervisor. When this occurs performance suffers and IT ops lose network visibility plus troubleshooting abilities. Look for a new market in VM networking to emerge to solve this problem.

3) Virtualization Is Systemic: While the industry has focused on data center virtualization and some networking suppliers communicate networking virtualization, it is clear that IT is being virtualized as evidenced by VM sprawl. Virtualization delivers two main benefits: expanding single IT assets to be available to many or to manage/pool many IT assets as a single resource. For example, a rack of blade servers can be managed as a single server. An application can be virtualized so that its image is available as a logical entity on many servers, increasing its availability. Storage area networks are virtualized in an attempt to more efficiently allocate compute storage, making storage available to many servers. Certain aspects of networking have been virtualized for nearly ten years, such as Virtual Local Area Networks (VLANs), which divide broadcast domains to service many applications, thus increasing performance.
What is becoming evident is that virtualization is redefining the boundaries between networks, applications and computing. Virtualization will fundamentally change the IT industry as these boundaries are not just technology demarcations but industry boundaries impacting industry suppliers and IT organizational design. While virtualization is taking place throughout IT, it's the data center that will show the way as to how the industry will change. While we're in the data center, congratulations to Cisco on winning best of show for their Nexus 7000 Series.

4) The Application Delivery Market Shifts: With Cisco out growing Riverbed in the network and application performance optimization market, plus Cisco winning best of show for their WAE 674 with WAAS v4.1 and Windows Server 2008, the market has clearly shifted from an appliance to a platform approach. Zeus Kerravala of the Yankee Group and I predicted that this market would consolidate as single purpose appliances gave way to more vertically integrated devices, which provide security, UC, mobility, networking, and computing services to branch office operations. The support of virtualization plus IP video is next with physical security IP video feeds flowing through these multi-functioned branch office devices.

5) Transitional Model for UC Needed: I'm having second thoughts on the wisdom of changing the name of IP telephony to unified communications or UC. IP telephony was well understood, its economics and value proposition solid. Then Microsoft entered the market and the industry broadly now refers to voice on an IP network as UC. Now UC is not well understood; its economics and value proposition are a work in progress. For example, transiting from a TDM network to an IP telephony network provides 15 to 50% total cost of ownership reduction, depending on initial conditions. UC is most often thought of as a single desktop/laptop launch point for a range of communications such as IM, voice, voice mail, videoconferencing, etc. UC is one part of an IP telephony solution.
Which brings me to the topic of a transitional deployment model needed for communications. UC is being deployed as an overlay to an existing TDM and IP telephony network. At a time when most corporations are scrutinizing expenses UC, deployed as an overlay, increases operational cost while IP telephony reduces it. To avoid this doubling up on operational spend, IT leaders should consider deploying a common SIP-based communications management device which can connect legacy TDM, IP telephony and UC clients to SIP and standard telephony trunking. This SIP- based transition strategy will simplify operations as it provides a common connection platform for all IP and non-IP voice communications. UC is not the end game in communications; it's a transitional technology to communications-enabled business processes (CEBP) and beyond.

6) UC Expands to Mobile Workspace: Many UC suppliers have announced Fixed Mobile Convergence (FMC) solutions which are the linking of fixed telephony end-points such as desktop phones and messaging with mobile devices; but a new level of functionality is emerging. Cisco announced a relationship with Nokia to deliver what they call Mobile UC. Mobile UC extends a professional's workspace from desktop/laptop to mobile smartphone by linking directory, presence, visual v-mail and ring management so that these communication services are common across fixed and mobile communication workspaces. Also with the use of dual mode antennas in the Nokia series smartphones one can initiate and receive calls via Cisco Communications Manager when in a building while using the mobile network when out of WLAN range. This announcement is the start of a new wave in UC that increases its usefulness to mobile devices, further improving both productivity and ROI.

7) The Intelligent Network Edge: Thanks to WLAN mobility, UC, PoE, IP video, increased bandwidth need and a host of other network services and requirements move to the edge of the network; wiring closet switches are being reviewed and replaced. New wiring closet offerings from Foundry and Extreme Networks highlight the increased demand for a new intelligent network edge which supports WLAN APs, efficiently delivers PoE, supports IP video and are UC port smart to acknowledge IP phones and ease their configuration. The change out in wiring closet switches has started as many IT leaders now realize that they need to increase the intelligence of their network edge to support the above applications and to lower operational cost. Many procured their edge network via the lowest cost provider resulting in numerous suppliers, high operational cost and an inability to scale and support new corporate applications.

8) IP Video Drives Convergence 2.0: With the massive consumption of IP video on enterprise campus and WAN networks, a new wave of convergence is poised to occur. With on-demand and videoconferencing IP video services doubling network traffic, IT and business leaders embrace convergence 2.0 which is the optimizing, scaling and securing of networks to support a host of IP video applications.

9) Green Networking: The networking and communications industries do more to reduce greenhouse gases and carbon emissions than any other industry. Communicating over large distances avoids travel while downloading documents, presentations and all forms of content avoids postal and express delivery. But every industry can and must optimize its products, energy consumption and exhaust. Green networking is primarily focused around efficient distribution of PoE, efficient power supply management of switches, routers and appliances and their associated cooling. The largest opportunity for a big green impact is in data center networking which is where all suppliers focused their investments and efforts. Foundry Networks was the winner of the Best of Interop Green Award for their green networking efforts in the data center.

When you add it all up, virtualization and communications continue to force a re-design in enterprise networking and IT in general. Virtualization, especially in the data center, and the integration of communications into an IP network, that being IP telephony and IP Video, are the two IT initiatives offering compelling ROI with reduced TCO plus increased functionality. As a result counting network ports will become increasingly meaningless over the next business cycle; the network edge needs to be more intelligent; and to connect branch office networks to data centers the application delivery market needed to change; all of which is being done via green networking.

2 Debates over Lippis Report Issue 105: What I Learned At Interop

  1. Art Rosenberg said:


    How off-base can you get about what unified communications is all about?

    I can’t really blame you for confusing “unified communications” with IP Telephony (IPT), because the biggest change in technology and business communications is taking place with the role of phone calls. Not only is the network infrastructure for voice telephony changing (VoIP), and the user interfaces moving away from the limitations of the TUI to richer, screen-based flexibility, but the use of more intelligent presence and availability and the seamless interoperability between all forms of messaging and information exchange with conversational voice contacts, will make for more flexible and efficient business operations.

    Voice telephony is no longer the only form of real-time communications, and it has always been a location-based connection, rather than a person or role-based form of contact. In business, the latter is becoming recognized as the real objective for real-time person-to-person contacts, as opposed to socializing. With presence-based IM, text messaging and information exchange have also become real-time, and but also allow escalation to the efficiencies of voice conversation once a successful contact has been efficiently established.

    Changing TDM telephony to SIP-based IP Telephony is a migration step towards unified communications, but it is not going to replace the increasing dominance and efficiencies of both real-time and asynchronous forms of messaging in business communications and information exchange.


    Art Rosenberg
    The Unified-View

  2. Art Rosenberg said:

    I should also point out that moving to applications-based contact initiation via “CEBP” is another step in the right direction of UC. This is where a business process needs to contact a person, and, believe me, its not with a voice conversation, but with messaging (with voice or text interfaces) and online interactions. The business process application may also intelligently coordinate a voice conference with other people when necessary, based on presence, availability, calendars, etc. This where the Internet and mobile devices will enable notifications (messages) to be delivered to people wherever they happen to be, along with real-time, contextual response links.

    So CEBP is simply an extension of UC in allowing applications to also efficiently contact people and deliver information in real-time through messaging. The problem that UC is really trying to solve is making contact with people as flexibly and time-efficiently as possible. Telephony alone could never do that!