Guiding Businesses Toward FULL NETWORK CONVERGENCE.
Few businesses will take the convulsive "forklift upgrade" path toward convergence. Most will seek a staged migration with the priority to squeeze the most from old network investments and assure that anything bought in the interim can still be used when they get there. Many will need help developing and executing strategies to maximize the existing equipment usage and minimize the amount of work habit change employees must endure to accommodate new systems.
Equipment vendors are eager to feed the appetites of MIS managers who will find irresistible the potentially huge cost savings of redirecting corporate telephony to packet data networks. New universal access port architectures can address the intersecting of voice/data convergence and remote access. Voice gateways between conventional PBX equipment and the IP network can maintain user familiarity. But it is the resellers, OEMs, and integrators dealing directly with end users--and who have their ears and trust--who must set the table for the convergence banquet.
Setting The Table
Before being presented potential solutions to the convergence challenge, businesses should do some homework. First, they should be made aware of what works today in the realm of IP technology and what is still evolving. For example, physically translating analog voice data into IP packets is old hat now with industry-accepted standards. However, sending real-time data like packetized voice over something like the "best-effort" Internet remains a work in progress with delivery and quality often erratic. Which applications would be appropriate for any individual company's "first stage" of migration? For large companies with sophisticated IT departments, this analysis is mostly an internal responsibility. For smaller enterprises, the mentoring role must fall heavily on outside partners.
Also up front, these businesses should have a comprehensive handle on their private network's operational capabilities--how it behaves, what it's designed to do, and what it actually can and cannot do--and how much "head room" is built in to accommodate growth in capacity. Available answers may vary depending on the size of the company and general philosophy of those running the IT department. Some outfits may be a bit frizzy if they're used to seat-of-the-pants solutions, like simply throwing bandwidth at problems. Others have very experienced people running carefully designed networks, which they understand completely.
A parallel assessment also needs to be made on just how able and willing their workforce is to deal with change at the desktop level and how much support and/or training may be needed to make a transition. Aside from the steep cost of a full-bore upgrade, another reason for taking the staged approach to convergence is that there will be less change for users to adjust to at any one time, so learning can be gradual. Again, trusted integrators, VARs, and OEMs can be instrumental in sorting through such questions, depending upon the nature and level of corporate IT resources available (and, quite often, the dynamics of internal corporate politics).
A short list of areas these businesses should revisit and comprehend thoroughly before launching conversion efforts should include:
* Network infrastructure - Depending on the current network's topology and capability, some generic items like hubs, switches, and routers may need reconfiguration or software upgrading to better handle QoS or for improved segmentation to handle the extra bandwidth.
* The Desktop - Is the telephone system adequate? Will individual PCs need hardware or software upgrades to handle IP multimedia applications? Is customer-premise type equipment used by "road warriors" (modems, etc.) adaptable for both data and voice?
* Network Transport - This is the time to re-examine WAN network services. What level of dedicated bandwidth and QoS can be guaranteed on current lines? Can current service levels meet the future needs for voice and fax over IP? If not, can the incumbent network transport provider upgrade service or should another provider be sought?
Why Universal Ports?
With the capabilities and needs assessments completed, companies plotting convergence can consider key technologies that traditional corporate telephony equipment vendors can offer to add voice-over-packet services with minimal migration costs--specifically voice gateways and universal network access devices.
Although some may opt to replace PBX systems and their conventional telephones with PC-based or standalone Internet phones linked directly to the IP network, a corporate PBX, by connecting to an external voice gateway device, can perform IP telephony functions without changing the terminals on workers' desks. This leaves undisturbed all familiar features of the office telephone system while adding (at negligible cost) the ability to route a number of calls over portions of the corporate data network. Alternatively, the PBX can incorporate the gateway functionality to create a direct connection to the IP network, but highly integrated modules from companies can do away with the need for PBX vendors to develop codec, network-delay compensation, or voice and fax packetization technology.
Presuming these businesses are insisting that anything they buy for their networks had better be able to support both voice and data, the ideal data network access device would service ISDN, data, fax, and voice calls on all dial-in ports. Further, dial-up ports ought to be scalable to meet traffic growth and shifting needs (i.e., the percentage of voice vs. data) and be software upgradeable to accommodate enhancements like new modulation or QoS standards. Such devices built on a universal access port architecture can also lower systems costs for OEMs, VARs, and integrators. They quadruple the number of ports per slot while using about half the power of standard data-only ports. This means not only increased functionality, but also lower power and cooling requirements.
More importantly, a universal port architecture is a future-proof investment. It already can see, hear, translate, and transfer every dial-up "voice" out there today--i.e., analog, digital, ISDN, voiceover-IP, etc.--and is software upgradeable to handle emerging new technologies. These next-generation dial-up ports, once known as modems, can accommodate any transmission protocol on any port in a network access server. Such gateway processors will make sense of an otherwise incomprehensible digital Tower of Babel being built on conflicting technologies and incompatible protocols.
Fully converged voice/data/fax/video networks will set the stage for untold new applications, which could ride these infrastructures. The wider it's deployed, the more entrepreneurs will be looking for ways to make money off them and new business models will be created. Between now and then, however, companies wisely guided by their outside partners should be able to preserve the value of their existing network equipment and use it effectively in the meantime, and choosing network access equipment that's scaleable, software upgradeable, and universal will make everyone took (and sound) good.
Mark Bernier is the director of product marketing at Mapletree Networks (Westwood, MA).
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Technology Information|
|Publication:||Computer Technology Review|
|Date:||Nov 1, 1999|
|Previous Article:||Sony DTF-2 Drive Zaps All Competitors.|
|Next Article:||Remote Network Management: The Make Vs. Buy Decision.|
|Metaphors converging on the Internet.|
|What Is Convergence?|
|NTT Com Announces Financial Results for Fiscal Year Ended March 31, 2006.|