Users and Suppliers Trying to Preserve Functionality of Private-User Networks.
Software-defined or virtual networking represents an alternative to private, dedicated networks that use special service circuits. In a true virtual network, there are no dedicated intermachine trunks and perhaps no dedicated access lines. Another alternative, however, is to combine a mix of public and private network services in a hybrid arrangement. The network and user features remain consistent and allow the customer to manage and control network usage.
At the present time, there are two distinct thoughts on the subject. The first is that software defined networks are provided by a carrier. The second approach is to provide the same capability through an existing PBX or a special applications processor. By utilizing a wide range of competitive services provided by multiple carriers, the network manager can meet user needs and exploit the advantages of competition.
Today's networks have evolved during the past two decades. In part they have been influenced by the specific needs of large and medium users, and in part they are the result of advances in technology. ICA member companies have been on the leading edge of these advances and often have run at a faster pace than the marketplace could respond. In most cases, AT&T introduced a network service in response to user requirements.
In 1963, AT&T introduced its first network offering that was known as CCSA. Using common controlled signaling with AT&T central offices, CCSA replaced manually intensive tie-line configurations. Then in 1976, after more than a decade of product evolution, EPSCS was filed. EPSCS offered the user better technology, more features, and enhanced management and control techniques. By 1979, electronic tandem networks emerged with intelligent PBXs or Centrexes providing the network switching vehicle. The latest network service offering comes in the form of SDN. The common thread to these offerings is the use of telephone company central office equipment and the prevailing tariffs that influence the user's network decision.
Equipment suppliers have long recognized that almost any service offering can be repackaged as a product offered directly to the user. More importantly, these offerings are usually tailored to meet specific needs more quickly and more economically than general purpose solutions. Such has been the case with networks.
The realities of the industry are actually quite simple. The trends that permeate telecommunications are digital technology, competitive long-distance services, an emphasis on public-switched facilities, carrier selection (as opposed to traditional facility selection), measured usage pricing, distributed control, and equal access. The combination of these trends is significantly changing the way we view networks.
Given this background, let's examine software-defined networks from a product perspective. First, what are product-based virtual or software-defined networks. Second, why have they emerged and can they be considered competition for carrier-based networks. Third, who's providing and promising it. Fourth, what type of features do these products offer users today and tomorrow. Fifth, when will they be available and more importantly, widely used. Finally, how much will it cost.
The distinguishing characteristic of product-based virtual networks is the location of the processor. Call processing in a carrier-based offering resides in the switch location of the long-distance carrier and provides interconnection with the long-distance services provided by a single source. Call processing in a product-based offering resides on the customer's premise and provides access to the most effective long-distance services provided competitive carriers.
Call processing intelligence, the database, is maintained centrally in carrier-based software-defined networks, whereas product-based systems contain multiple databases tailored to a specific location. A centralized database is relatively easy to maintain and product manufacturers provide centralized management capability. In a distributed network, each database can be tailored to meet the unique characteristics of a location.
Gaining access to private lines or existing private-line networks is another differentiation. Carrier-based systems require specific private network gateways or PBX software that controls the call processing locally. Product-based virtual networks integrate the use of private lines into their approach. Take the example of the "lazy tie line." The first two or three lines in a typical four or five-line network trunk group are extremely cost effective, while the last two lines maintain grade of service. The product-based virtual network packs the first two lines with traffic, then overflows to "virtual tie lines" during peak busy hours while maintaining the same grade of service transparent to the user.
Both examples point out the hybrid type of a virtual network that can be implemented. By taking the best of both worlds, the user can still take advantage of economical private-line routes, while exploiting competitive switched network services. This arrangement provides a great deal of short term and long term flexibility.
Perhaps the most significant distinction between carrier and product approaches lies in the ability to select among competitive services. Each carrier service is designed and operated on the assumption that all traffic comes directly to the carrier's nearest point of presence. A product, on the other hand, permits the network manager to route each call to the most effective carrier, based upon predetermined route structures and the requirements of the user, whether it's lowest cost or best quality.
Another characteristic of the product-based software defined network is the call information management system. System and usage statistics, as well as individual call accounting, is a key ingredient of network management and control. In a product environment, a call record can be created for each originating and terminating call attempt. A centralized polling device can be used to collect the information from each site for further processing.
Briefly, a product-based virtual network provides the same feature set and networking capability of a carrier-based software defined network, but permits the network manager to tailor the system to meet the specific needs of the user.
Networks represent a delicate balance between its physical ingredients--switching, transmission, and signaling--and the user perspective: quality, grade of service, as well as management and control.
Software defined or virtual networks have emerged because they represent the next logical step toward the concept of integrated services digital networks or ISDN. The most crucial issue of ISDN is who controls the service node. Carriers want to provide the intelligence to interface with various services in order to collect the revenue. Equipment manufacturers want to provide the intelligence at the customer's premises so they can sell hardware and software that selects among alternative services. This struggle to control the user's destiny is classical.
Virtual networking is seen, at least by suppliers, as a competitive response to rising private-line rates as well as extended service provisioning delays. The challenge for network managers is still the same--provide a cost-effective method of transporting information within the company to its supplier community and its customers.
Virtual networking has emerged because it is another "window of opportunity" to capture a piece of the lucrative large user communications market. Software-defined networks have emerged because technology allows it and regulatory events have drawn us toward it. Virtual networking can provide an effective pathway to future networks whether it's ISDN or something else.
The future and trends of software defined networking features will tend to migrate toward ISDN capabilities.
Some will consider these standard, others will charge by the feature.
How soon will virtual or software-defined networks be coming to your neighborhood? Soon. The consensus seems to be that 1985 is the year for introduction, beta testing and field trials, and general introduction beginning shortly thereafter. Carrier-based systems are generally all-inclusive and often require longer lead times from field trial to general availability. On the other hand, product-based systems, particularly attached processors, can be deployed location-by-location.
No one appears to be far-ahead of the field at this time, although Isacomm has announced that 30 users will be on board by the end of 1985--one is currently installed. General availability will occur during 1986 and 1987, whereas the five-year period to follow will result in a proliferation and general use of virtual networking. During this period, users can also expect the first generation of ISDN capabilities to arrive on the scene.
What's It Cost?
Will virtual networking save the network manager any money? It depends on what he's doing now. Estimating the cost, even putting a ballpark number, for a total network is difficult to do. Keep in mind that the numbers you see are rough estimates based on assumptions for the "typical" network. Which means, if you want a more specific proposal, see your vendor.
First, let's see what carrier-based SDN will cost. As it is filed today, the cost for SDN transmission ranges between 11 cents and 38 cents-per-minute. A network which includes a full feature set, 50 percent off-net traffic, and 70 percent of all locations attached by standard access lines, will average out at between 27 and 35 cents-per-minute.
Product based systems, on the other hand, require up-front capital and then incorporate competitive switched services costing between 27 and 35 cents-per-minute.
Hybrid systems, on the other hand, will exploit the presence of some private lines--particularly as the lead lines of a trunk group--to bring the cost-per-call minute down to the range of 15 to 25 cents-per-minute.
Naturally, all of these numbers will vary with the traffic patterns of each company and each location. What we have found in our network analysis at COM, however, is that the hybrid arrangement seems to be the most viable solution given the tariffs that exist today. Total SDN may increase the overall cost of the network by eliminating cost-effective private-line routes and total product-based solutions rely too much on the availability of cost-effective competitive long distance services.
In conclusion, there are a few other questions that we should try to answer today. First, what's it going to do for the telecommunications network manager? It will provide a cost-effective way to revive sagging network economics and extend corporate network services to marginal locations. Second, what's it going to do for the individual network user? It will provide them with the advantages of equal access, competitive long-distance services and the economics of private lines, and WATS without the confusion--user transparency.
Also, what's it not going to do? It will not relieve the telecommunications manager of the responsibilities associated with managing a corporate network. If you want to save money with a corporate network or get the respect from your users by providing quality service, you've got to do it the old fashioned way--you gotta earn it!
Is the concept of software defined or virtual networking for your company? Right off hand, we can't answer that question and most likely you can't either.
Carrier-based solutions will provide the network manager with a relatively simple, single-source network. This can be very attractive to users with small staff allocation or minimal internal support for capital expenditures.
Product-based solutions may fit you if you want to exploit equal access and competitive long-distance services. Since many users are upgrading local telephone systems, the feature sets offered by PBXs or attached processors may solve other concerns such as equal access connectivity, call information management, or individual user control.
On the other hand, however, users have already invested heavily in private-network switching and transmission facilities. Many have been barely "hanging on" since the 1981 rate increases exploded private-line network economics. The "no network" alternative has led many companies to begin dismantling the corporate network. Virtual networking may offer yet another alternative for those who want to preserve networking throughout the company.
|Printer friendly Cite/link Email Feedback|
|Date:||Jun 1, 1985|
|Previous Article:||Continued Changes in Communications Point to Growing Telemarketing Needs.|
|Next Article:||How Network Analysis-Management Software Supports the Telecommunications Manager.|