ISDN; Frame Relay; ATM
In WAN as in most areas of business the desire to maximise the impact of any investment in technology is a core issue. Careful analysis of data traffic as well as business objectives will help to successfully match business applications with the networking services.
Some Business Considerations for choosing systems:
- How affordable is it?
- How complicated is it?
- Is it transparent?
- The capacity for true bandwidth on demand.
(The greater the bandwidth indicates faster or greater data transfer capability)
ISDN (Integrated Services Digital Network)
- ISDN is a switched digital network service that accommodates voice, data and imaging to other ISDN end users.
- It converts analog telephone lines to digital.
- Its goal is to link homes and businesses over telephone wires.
- Sometimes known as narrowband ISDN.
- Local Loop Transmission Technology.
- Around for 10 years but is only widely used in last 3 to 4 years.
- Not designed to be 24hr like T-1 or bandwidth on demand like frame relay service
- Main problems are cost current availability of infrastructure.
ISDN is deliverable on two service levels:
Basic Rate Interface (BRI) – also called 2B + D
- The ISDN basic rate interface is intended to replace standard analog telephone services.
- It is designed for use on a dial up basis.
- The BRI has 2 Bearer channels (B channels) at 64 Kbps each – these channels transmit user data (voice, video and data). There is also a Data channel (D Channel) at 16 Kbps – the D Channel is for carrying network-signalling data.
- Together the basic rate interface makes up a bandwidth of 144 Kbps.
- The use of the D Channel for carrying signalling data that is not intermingled with the user data is known as "Out-of-Band Signalling".
Primary Rate Interface (PRI) – also called 23B +D
The PRI can have 23 x 64 Kbps B Channels and a 64 Kbps D Channel. These are the minimum rates but they can be greater. This gives a bandwidth of 1.544 Mbps. With additional overheads businesses could consider mapping this onto the T-1 circuit because PRI has the required bandwidth to make use of T-1 transmission rates.
[The T-1 circuit is a point-to-point transmission technology with high-speed digital lines that use two wire pairs (one to send and one to receive). Other countries use a similar service called E-1 (European Digital Signal 1) – signalling rate of 2.048Mbps.]
Note: One of the above rates does not have to be used exactly. You can increment your ISDN capability rather than jumping straight up to PRI from basic rate interface. Multi-rate ISDN uses the inverse multiplexing technique which is a collection of 64 Kbps B-Channels dialled up together into a single logical channel to meet the application needs e.g. videoconferencing can be used to attain variable amounts of bandwidth.
ISDN has been criticised for only having isolated deployment of the infrastructure in many regions but the coverage situation has improved over the past 3 to 4 years with more its widespread use.
Considerations and Hindrances:
Distance: Signalling requirements can limit ISDN installations, as they must be within 3.4 miles of an ISDN switch.
External AC Power Supply: Unlike today’s analog phone lines premises with ISDN require an External AC Power Supply because it is digital. There is some concern of losing the ISDN-based phone service during power failures whereas today’s analog phone lines would not be affected.
Charges:
- ISDN charges vary between carriers – flat monthly rates and/or per minute usage charge.
- Role of Thumb for cost: ISDN access for office to office connectivity is cheaper than leased lines of equivalent bandwidth when connectivity needs are for 4 hrs/day or less.
- For occasional communications between offices for LAN-to-LAN connectivity or database updates, ISDN routers can be much more cost effective than leased line routers.
Incompatibility: ISDN switches have had slightly different specifications but elimination of compatibilities is taking place with interoperability specifications developed through ‘National ISDN-1’. This defines a national standard for ISDN switches and inter-switch communication.
ISDN Ordering and Implementation: To properly link end user’s equipment such as remote access servers must be programmed with service profile identifier numbers to properly identify the carrier’s equipment must link (must interface).
Uses of ISDN
- Single user-to-office
- Single user Internet connectivity
- Office-to-office connectivity
Applications for Single User
- Telecommuting
- Internet Access
- Simultaneous voice and data technical support
- Collaborative computing
Applications Office-to-Office
- Remote office routing
- LAN to LAN connectively
- Disaster Recovery with ISDN lines as backup to failed leased lines
ISDN Terminal Adaptor
It is the equivalent of an ISDN modem and puts voice or data onto the ISDN. It allows analog devices such as phones and fax machines to interface to all digital ISDN. Software and drivers must be compatible with installed network operating systems as well as with the purchased ISDN terminal adapter.
Network Terminator Unit –1 is required to physically connect the ISDN line to the user’s ISDN equipment.
Broadband ISDN (B-ISDN):
B-ISDN delivers voice, data and image from the joint network architecture of ATM and SONET. SONET is the optical transmission interface and mechanism that will deliver B-ISDN services. B-ISDN should be the service
Frame Relay
- Frame relay is a fast packet switching technology.
- It is a point-to-point system that transmits variable length frames at the Data Link layer through the most cost effective path.
- Requires frame relay routers or bridges.
- Network Service
- It is a Switching Architecture
- Primarily for the purpose of LAN interconnection
Technology
Advantages
- Frame relay involves two layers of the protocol stack - the physical and data link layers. There is a decreased processing time and an increased throughput time because the network layer is not involved.
- Frame relay networks often employ permanent virtual circuits (PVC’s) to send frames from source to destination through the frame relay cloud.
Multiple PVCs can exist within frame relay clouds and provide the advantage over circuit switched options that the multiple PVCs can be supported from only one access line
It was introduced to take advantage of the decreased bit error rate on today’s transmission systems. By making this assumption point-to-point error correction and flow control have been eliminated from within the network itself. Instead the end nodes look at any problems encountered along the way. Thus data is transferred faster.
Errors are still detected but the frame relay switches do not try to correct the bad frames but instead discard them. The correction and request for retransmission of bad frames is left to the end node devices of modems and computers.
Disadvantages
- If there are errors they take longer to correct, as the errors have to be retransmitted form the receiving node to the initial sending node and back again. The frame relay network has transmission speeds of T-1 (1.544Mbps) and occasionally T-3 (44.736Mbps).
There is no way for end nodes to either monitor or manage flow control within the frame relay network itself. Only end nodes can manage flow control between themselves and whatever frame relay network access device they are linked to.
- Does not operate within a LAN whereas ATM does.
- The lack of predictable maximum delivery delay is a key weakness in frame relay.
Business
Advantage
Unlike the uniform cells of ATM, Frame Relays transmit frames or packets of variable length with a maximum of 8,000 characters. The potentially large frames also create high throughput and low delay.
Disadvantage
As there is no defined length to a frame there is no guarantee as to how quickly a frame can be forwarded. This is only an issue with time-sensitive information such as digitised voice and video. For this reason if a business makes frequent use of either of these media, ATM is recommended instead. As a result of this drawback of variable length packets, frame relay is often described as a data-only service. Data frames need a lot slower end to end connectivity.
- Data is transmitted through the frame relay cloud along the fastest and most cost effective route. This transmission is transparent to the end user.
- A Committed Information Rate is supplied to businesses on leased lines. This refers to the minimum bandwidth guaranteed to users for normal transmission. When leasing a line the business can agree with the vendor (e.g. Eircom) on the information rate of e.g. ½ Mbps At some point the bandwidth requirement might rise and then there is the option available to pay for the additional requirements. This is cost efficient as the business only uses extra bandwidth when required.
- If cost were a big issue frame relays would be preferential choice over ATM if the business were only likely to be handling heavy bursts of traffic on dispersed occasions. The dynamic bandwidth allocation will handle this by assembling and forwarding more frames per second onto the frame relay network and over multiple permanent virtual circuits.
- Inter-LAN communication tends to have these large bursts of requests for data and file transfers.
- There is need for better coordination among different frame relay vendors in order to offer more transparent access between them.
- Frame relays are protocol independent or protocol transparent. What goes on inside the frame relay network cloud remains transparent to the end user.
ATM (Asynchronous Transfer Mode)
- ATM is a Broadband cell relay method – Therefore also known as cell relay.
- It is an advanced form of packet switching.
- The WAN technology of the moment for businesses.
- Very High Speed (155 Mbps to 622 Mbps) e.g. can transmit the entire Encyclopaedia Britannica in one second.
- There are variable rates of transmission but much faster than frame relays.
- Multimedia technology that allows WAN to have simultaneous transmission of voice, video and data.
Technology
- ATM networks have fixed length cells of 53 octets long (53 bytes) that are small and move very fast. 48 bytes are for data or information from higher-level protocols and the remaining 5 bytes for header information. It breaks packets down into a smaller size but doesn’t need much headers and footer data. ATM is a lot faster than frame relay switches because of the uniform cell length.
- Also provide predictable delivery times that frame relay cannot offer.
- ATM adaption layer (AAL) in the data link layer processes the input information into the fixed-length ATM cells so an ATM switch can send them on. The switch routes the information to other locally connected ATM devices as well as the wide area ATM network. AAL are designed to optimise the delivery of a wide variety of types of traffic.
- ATM protocols are supported from the LAN to the WAN, from network interface cards to ATM WAN switches, thereby removing the necessity for multiple protocol conversions from the desktop across enterprise networks.
- Just have general idea of packets in ATM
- Relies on carriers to implement ATM on WAN, leasing off local lines.
- ATM switches act as hubs and/or high-speed routers to remote networks.
- Permits several computers to transmit at once but its system works like an ambulance in traffic, giving priority to voice, visual and data.
Business
- ATM has LAN and WAN functionality but frame relay has only WAN functionality.
- If business makes substantial use of multimedia it has to be considered over frame relay.
- Very expensive because it requires special hardware and exceptionally high speed bandwidth to reach its potential. This bandwidth is not yet fully available to support full speed ATM in a WAN environment.
- Installing requires extensive equipment replacement.
- The availability of components can sometimes be a problem as there are currently only a limited number of vendors.
- ATM will require considerable expertise.