Most ISDN installations in remote locations use the Basic Rate Interface (BRI), offering users two B channels for user data and a single D channel (16Kbps) for signaling. This provides a total bandwidth of 144Kbps; however, each B channel is only 64Kbps, for a total user bandwidth of 128Kbps.

	ISDN BRI is really a 192Kbps circuit; the remaining bandwidth of 48Kbps is overhead. The physical frame in ISDN BRI is 48 bits, and the circuit sends 4,000 frames per second.

Host connections typically terminate with ISDN PRI (Primary Rate Inter-face) services, which use T1 circuits. This provides 23 B channels, and all signaling occurs on the D channel. Each channel is 64Kbps, for a total data rate of 1.535Mbps. The remaining bandwidth is overhead.

Designers should carefully review the costs associated with ISDN before committing to the technology. Since most tariffs are based on per-minute billing, bills in the thousands of dollars per month are not uncommon when improper configurations are deployed. This factor is the largest negative regarding ISDN for telecommuting. Users will also notice that connections require a few seconds to be established—ISDN is not an always-on technology.

	A D-channel-based service, called always-on ISDN, is available from some vendors. This provides up to 9.6Kbps for user data and can be used as a replacement for X.25 networks.

Communications over ISDN may use the Point-to-Point Protocol (PPP) where desired. PPP provides many additional services, including security via the Challenge Handshake Authentication Protocol (CHAP). PPP is an open standard defined in RFC 1661, and the PPP protocol, through the Link Control Protocol (LCP), performs initial configuration. Multilink PPP may be used for B channel aggregation as well.

Multilink PPP (MP) performs a number of functions, but it primarily is responsible for the segmentation and sequencing of packets across multiple channels. This bonds the two B channels for a total of 128Kbps of user data, but it does not allow each channel to cross multiple chassis. The protocol is defined in RFC 1717, and it adds four bytes of overhead to each packet on the link. Network designers may find this function useful in videoconferencing applications; however, it is also applicable for remote data connectivity.

The Multilink Multichassis PPP protocol (MMP) is another protocol that combines B channels. MMP operates across multiple routers and access servers and is more scalable than the standard Multilink Protocol. Various B channels can span numerous chassis, allowing for larger, more scalable access farms and better redundancy options, since the failure of a single switch may not result in the loss of a session. When additional capacity is needed for a cluster, an administrator need add only another peer access device.

MMP relies on an MMP process server to reassemble the calls. One possible implementation of this would include a 4700 router fronted with multiple AS5200s. The AS5200s combine to create a logical federation called stack-group peers, and these peers use the Stackgroup Bidding Protocol (SGBP) to elect a process server. SGBP is a proprietary protocol. Although MMP may be used similarly to MP, the multichassis nature of the protocol allows for greater scalability and aggregate bandwidth. The SGBP process selects resources based on previously existing sessions and processor load.

ISDN may also be used for L2F (Layer 2 Forwarding Protocol), PPTP (Point-to-Point Tunneling Protocol), and L2TP (Layer 2 Tunneling Protocol) tunnels, which are described in Chapter 11. These secure conduits are ideal for Internet connectivity; however, they may also be used in private networks. One application for tunnels includes telecommuting—rather than having all users call a central, long-distance number, they can call a local point-of-presence and pay for a local call. The point-of-presence may be private and be maintained by the corporation or an ISP on the Internet. This concept is used for Virtual Private Network (VPN) solutions.

Remote Access

Over the years, users have demanded access to corporate LANs from their homes, hotel rooms, and customer sites. These requirements depart significantly from the fairly comfortable and controlled structure of the local area network.

In fact, many companies have decided to address remote connectivity with VPNs or with a combination of services that are outsourced to a provider. Outsourcing is a good way to control costs, although the costs are generally higher than with internally administered remote access solutions. This setup works in most corporations because hiring full-time personnel is very costly. Frequently, outsourced solutions can also decrease communications costs, which are recurring and can quickly overrun the best budgets, as the major telecommunications providers maintain points-of-presence in virtually all calling areas. For the corporate user, the call into the remote-access system is a local one rather than a long-distance or 800 call, each of which costs the corporation substantially more.

Network designers need a thorough understanding of the remote connectivity options for either outsourcing or corporate-provided solutions. These solutions incorporate remote nodes, remote gateways, and remote control. However, this text will also incorporate remote users and their requirements into the mix.

It is important to note that most of these solutions are still deployed on standard telephone services, although some deployments use ISDN. Within the first few years of the 21st century, it is likely that cable modem and xDSL solutions will also be incorporated into remote-access deployments, and these technologies will likely replace ISDN and POTS.

Designers need to realize the limitations that come with any of these transport technologies. For example, standard telephone services are slow, but they are also universally available. Solutions based on DSL are fast and comparatively cheap compared to ISDN and analog dial-up (based on bandwidth), but they must be pre-installed and are fixed in location at the remote end. While this makes the higher speed solution less attractive to remote users who travel, it would be an ideal solution for an at-home telecommuter.

Remote Gateway

Remote gateways are designed to solve a single remote access need, and as a result they can be fairly inexpensive. The most common remote gateways are used for e-mail, but they can be configured to provide other services as well. A remote gateway is a remote-access device that services a single application.

The key to remote gateway solutions is that they generally do not scale because the remote gateway device typically processes the application in addition to the remote session. Therefore, the designer may address a single need quickly without building in scalability. As a result, the designer selecting remote gateway technology would likely purchase separate modems and phone lines for each gateway deployment—resulting in an expensive long-term solution as more and more gateway services are added.