Task 1

1. Delineate the distinguishing attributes of a WAN.

Answer:

Communication between computers located at different sites requires the involvement of a public communication service such as the local and long distance telephone companies or other providers. An exception is the campus network, which interconnects networks between buildings in a campus or industrial park setting, and in which the owners can deploy their own cable systems among buildings.

There are a number of wide area connection options as follows. Modems are employed when analog telephone lines are used. They convert computer digital signals to analog signals. Digital lines are also available from the carries that provide faster and more reliable service as either dedicated or dialup lines. A dedicated line is a permanent connection between two pints that is usually leased on a monthly basis. Any-to-any connections provide a way for users to connect with a number of different sites, not just one dedicated site.

· A switched point-to-point connection using modems. the line is connected only during the call.

· A dedicated point-to-point connection over analog lines using modems.

· A switched point-to-point digital line.

· A dedicated point-to-point digital line.

· Any-to-any digital lines over circuit-switched networks.

· Any-to-any digital lines over packet-switched networks.

Modems are not required on digital lines. Instead, a channel service unit/data service (CSU/DSU) device provides the connection between local equipment and data communication network. These devices are called data communication equipment (DCE). DCE devices sit between data terminal equipment (DTE) devices, such as computers, and the communication network.

Other methods for connecting systems over wide geographic areas include microwave and satellite communication systems. Microwave systems consist of transmitters on top of earth-based towers that transmit signals over large global areas.

2. How is communications accomplished with circuit switching technology?

Answer:

A circuit is a path that can carry signals from one device to another. The most common analogy is the telephone. When you call someone, a dedicated circuit is established between two telephone devices. The telephone system is a circuit-switched network because the local telephone company provides the switching capabilities to connect your telephone with many other phones. Circuits are set up for the duration of the call and discontinued when the call is complete.

In the context of computer networks and data communication, there are two types of switched data communication circuits provided by the telephone companies:

· Analog Telephone Lines. These are dialup switched circuits that provide temporary connections using modems (modulator/demodulator). Modems convert digital signals to an analog signal for transmission over analog telephone lines. A modem is required at each end of the connection and the speed of the modem determines transmission rates. They are usually inappropriate for real-time access to remote resources by multiple users.

· Switched Digital Services. These provide more flexibility than leased dedicated lines. They are basically dialup lines that provide better quality service than comparable analog switched lines and are more suitable for LAN-to-LAN transmissions. Carriers provide switched digital services under special contract because the lines must handle higher data-rates than customers would normally transmit over analog lines. The switched services have a big advantage over dedicated lines. They provide route flexibility so customers can get high-quality, high-speed digital services between any two points when it’s needed. In contrast, dedicated lines are set up between two specific points and do not accommodate moves. In addition, customers often end up paying for more than they need, then do not have enough service during peaks in traffic.

A circuit-switching service type of technology provides a temporary dedicated path through a carrier’s switching systems. Customers can contract for various types of services, depending on their anticipated bandwidth needs. For instance, the following are types of circuit switching technologies:

· Switched-56 services. This is a common switched service which operates a 56 Kbits/sec and requires a special Switched-56 channel service unit/data service unit (CSU/DSU) at each site. Switched-56 services were originally intended to provide an alternate backup route for higher-speed leased lines such as T1.

· Integrated Services Digital Network (ISDN). This is a service that provides all-digital services on the local loop, which is the cable that runs between the home or business user and the LEC’s switching office.

· Emerging services. Some emerging standards will boost rates and lower prices on the local loop. High-bit-rate Digital Subscriber Line permits transmission over existing copper-based lines between 784 Kbits/sec to T1 rates of 1.544 Mbits/sec. A related product, Asymmetrical Digital Line works over twisted pairs, offering a 1.544 Mbits/sec circuit in one direction. A third encoding scheme, Very High bit-rate Digital Line, will provide 3- to 6-Mbits/sec service over two twisted pairs.

3. Delineate key benefits and limitations of a virtual private network for information transport from the perspective of your organization.

Answer:

My current organization is using a virtual private network. In general, organizations can build private networks or public networks. A private network consists of switching and communications equipment owned by an organization that is interconnected with leased or privately owned communication lines (such as microwave systems). In a private network method each site will need three dedicated (leased) lines and associated routers to connect with every other site, for a number of dedicated lines and routers. The private networks use private or leased lines. A leased line is a communication circuit that is set up on a permanent basis for an organization by the phone company. Leased lines bypass switching equipment at the local exchange carrier (LEC) and therefore do not require a setup phase before each data transmission. Therefore, they are always connected. If the line is long distance, then the long-distance carrier is involved, as well as the LEC at the other end of the circuit. Some alternate carriers, such as MCI, are able to provide LEC bypass facilities in various metropolitan areas. Because the organization pays a fixed rate for the lines under contract, the lines are often called leased lines.

In my organization, or in general as a matter of fact, the key benefit or advantage of a leased line is that its quality can be verified and guaranteed. The line encompasses the local loop, which is the wire that connects a business with the LEC, and this wire is conditioned to ensure quality. Thus, it is possible to use higher data transfer rates on the line.

As above said, the leased lines used in my organization are used to build private networks, in which an organization interconnects its remote sites using its own switching equipment and takes advantage of the privacy inherent in and dedicated bandwidth of leased lines. The other key advantage of this is that with leased lines and private switching equipment, and organization maintains security and control over the traffic that crosses the line. However, building and managing private networks can be expensive when compared to using the alternative public networking schemes like X.25 and Frame Relay. A private network requires a dedicated line between each site, with a bridge and/or router at each end, so to interconnect four sites you would need a total of six leased lines. Using a public network, only one line into the carrier’s switching network is required from each site.

The carrier community is deploying the technology that organizations need to create virtual data networks among remote sites in the campus environment, metropolitan area, wide area, or on a global scale. A virtual data network provides a way to fully interconnect all systems at all sites in the organization at much lower costs than leasing dedicated lines and building private networks. The goal is to provide LAN-like speeds over area networks.

4. Indicate the main characteristics of an intelligent data network.

Answer:

An intelligent data network is a classical data network that offers superior capabilities such as management function, better data routing schemas (datagram), and flexibility in data formats. A hub is an example of one the components of such an intelligent data network. Hubs are central wiring concentrators that provide repeater functions in networks such as ARCNET and Ethernet 10Base-T. The hub serves as a central place to connect workstations and more easily manage the network. The first hubs were simple repeaters that supported a single transmission medium only. The wiring configuration they supported was appropriate for department or workgroup local area networks (LANs) of about 20 users. These hubs are called first-generation hubs, and they are still viable products for small LANs.

Second-generation hubs are called intelligent hubs because they include management features, such as the ability to detect faults, and collect information about network activities and the individual ports on the hub. The information is collected and reported back to a central management station. Simple Network Management Protocol (SNMP) management features are supported in most second-generation hubs. Other important features of intelligent hubs are:

· They include backplanes with multiple buses to support different media such as Ethernet, token ring, and FDDI.

· They typically use high-performance RISC processors that improve packet throughput performance.

· They have the ability to create logical LAN segments within a single hub and bridge those segments.

· They may have installable management modules that provide the ability to manage the hub from a remote location.

· They may have out-of-band signaling that connects remote management stations to the hub via a separate line that remains live even if LAN communications fails.

Third generation hubs have began to appear. These include more sophisticated management and monitoring features, faster backplanes, improved bridging and routing among segments, and the ability to microsegment the network so that a single LAN supports a single workstation. Emerging hubs provide Asynchronous Transfer Mode (ATM) switching between any ports.

5. Discuss the significance of the X.25 standard.

Answer:

The X.25 protocol is a CCITT (ITU) recommendation that defines connections of terminals and computers to packet-switching networks. Packet-switching networks route packets of data through a network to destination nodes. X.25 is a well-established packet-switching service traditionally used to connect remote terminals to host systems. The service provides any-to-any connections for simultaneous users. Signals from multiple users on a network can be multiplexed through the X.25 interface into the packet-switching network and delivered to different remote sites. A communication channel called a virtual circuit connects and stations through the network over a predefined path. The X.25 interface supports line speeds up to 64 Kbits/sec, although a major portion of the throughput is error-checking overhead. The CCITT revised the standard in 1992 and boosted the speed to 2 Mbits/sec.

X.25 is a standard, well-tested, and often revised protocol that has been a workhorse packet-switching service since 1976. It is suitable for light loads and was commonly used to provide remote terminal connections to mainframe systems. X.25 packet-switching networks are not suitable for most LAN-to-LAN traffic because they are slow and require a large portion of the bandwidth to handle error checking. This error checking was important in the days of low-quality analog telephone lines, but is not needed today. Frame Relay provides a better choice.

X.25 generally refers to the well-established packet-switching services that were traditionally used to connect remote terminals to host systems. X.25 defines a set of protocols for connecting to a packet-switching network. These networks use extensive error checking that was required in the days when telephone lines were not as reliable as they are today. A message is returned to the sender when a switch detects a corrupted packet, so the sender knows as soon as possible when there is a problem and it must retransmit a packet. This overhead reduces performance and is largely responsible for the growth of Frame Relay, which eliminates the error checking feature. X.25 operates in the 600 bit/sec to 64,000 bits/sec data transfer range.

Task 2

1. What are the distinctive characteristics of a LAN? Provide some representative examples of local network applications in your workplace.

Answer:

A local area network (LAN) connects computers in a workgroup, department, or building. In contrast, an internetwork is a collection of LANs within a building, group of buildings, or campus area, and a wide area network (WAN) is an internetwork that spans geographical areas and requires public or private communication links to interconnect each area. Usually a LAN is a communication network that is confined to a small area, such as a single building or a small cluster of buildings. The design of a LAN is determined primarily by the three factors: transmission medium, topology, and medium access control (MAC) technique.

All the companies I worked for the past 15 years employed a LAN. In some cases this was a part of a larger network, usually a MAN or a WAN. In my current workplace, we use an Ethernet-, coaxial cable-based LAN. On this LAN there are a variety of computers: PCs, MACs, and UNIX Stations, as well as resource shared devices such as printers and.

2. With the aid of diagrams, describe the key features of the following network topologies:

a. Bus

b. Ring

c. Star

Answer:

The architecture of a network is defined by its topology and the cable access method and communications protocols it uses. Before any workstation can access the cable, it must establish communication sessions with other nodes on the network. The cable access method of a network defines how a workstation gains access to shared media so it can transmit information. Protocols are the rules and procedures that systems use to communicate with one another over the network.

You can think of a network's topology as a map of its cable layout. Topology defines how you run the cable to individual workstations and plays an important part in the decision you make about cable. A network can have a LINEAR (BUS), RING, or STAR topology. You must consider the topology of a network when making decisions about which network type to install. Topology corresponds to how you will run the cable through the walls, floors, and ceilings of your building, as described here:

• Linear (bus) topology. A linear topology consists of a single cable that extends from one computer to the next in a daisy-chain fashion. The ends of the cable are terminated with a resister. Ethernet coaxial networks use linear topologies. While easy to install, a break anywhere in the cable disables the entire network.

• Star topology. In a star topology, all wires branch from a single location, such as a file server, or a central wiring closet. Star topologies require a cable to each workstation, but a broken cable only disconnects the workstation attached to it. Ethernet 10Base-T and token ring networks use star topologies, although internally, token ring is a ring network.

• Ring topology. In a ring topology, the network cable connects back to itself and signals travel in a ring. Physical ring topologies are rare. Token ring and ARCNET are logical ring networks.

3. Briefly compare the digital PBX and LAN in terms of installation requirements and reliability of information transport.

Answer:

The digital PBX is marriage of tow technologies: digital switching and telephone exchange systems. The forerunner of the digital PBX is the private branch exchange (PBX). A PBX is an on-premise facility, owned or leased by an organization, which interconnects the telephones within the facility and provides access to the public telephone system. Typically, a telephone user on the premises dials a three- or four-digit number to call another telephone on the premises, and dials one digit (usually 8 or 9) to get a dial tone for an "outside line," which also the caller to dial a number in the same fashion as a residential user.

The original private exchanges were manual, with one or more operators at a switchboard required to make all connections. Back in the 1920s, these began to be replaced by automatic systems, called private automatic branch exchanges (PABXs), which did not require attendant intervention to place a call. These "first generation" systems used electromechanical technology and analog signaling. Data connections could be made via modems. That is, a user with a terminal, a telephone, and a modem or acoustic coupler in the office could dial up an on-site or remote number that reached another modem and exchange data.

The "second generation" PBXs were introduced in the middle 1970s. These systems use electronic rather than electromagnetic technology and the internal switching is digital. Such a system is referred to as digital PBX, or computerized branch exchange (CBX). These systems were designed primarily to handle analog voice traffic, with the code function built into the switch so that digital switching could be used internally. The systems were also capable of handling digital data connections without the need of a modem.

The "third generation" systems are touted as "integrated voice/data" systems, although the differences between third generation and upgraded second generation are rather blurred. Perhaps a better term is "improved digital PBX."

Some of the characteristics of these systems that differ from those of earlier systems include: the use of digital phones, distributed architecture, and nonblocking configuration.

As new features and technologies are employed, incremental improvements make difficult the continuing classification of PBXs into generations. Nevertheless, it is worth noting recent advances ion PBX products that, together, might be considered to constitute a fourth generation: Integrated LAN link, Dynamic bandwidth allocation, Integrated packet channel.

Some of the advantages of the digital approach are: Digital technology, Time-division multiplexing, Digital control signals, Encryption.

And finally, it's worth to mention that digital PBX, with all of the above described features, added a great deal of reliability of information transport for LANs.

Task 3

1. What are the distinguishing features of ISDN?

Answer:

ISDN (Integrated Services Digital Network) is often proclaimed as the public telephone and telecommunication interface of the future, although it has been slow to emerge. ISDN integrates data, voice, and video signals into a digital (as opposed to analog) telephone line. The important point is that it brings digital services all the way to the home or office. While most telephone companies have already switched to optical cable and digital transmission for intra-city and inter-city links, the "local loop" that connects many home and business users to the telephone company switching office still uses signaling techniques. ISDN also standardizes the services provided to subscribers on an international level, thus providing a way to create international networks.

A number of small-scale ISDN facilities have been developed, but the promised extension of ISDN to encompass worldwide public telecommunications is some years off. Meanwhile, the characteristics of ISDN are defined by an evolving set of standard (being developed on a truly massive scale, both in terms of content and participants).

Activities currently under way are leading to the development of a worldwide ISDN. This effort involves national government, data processing and communications companies, standards organizations, and others. certain common objectives are, by and large, shared by this disparate group. The key objectives are:

• Standardization

• Transparency

• Separation of competitive functions

• Leased and switched services

• Cost and switched services

• Cost-related tariffs

• Smooth migration

• Multiplexed support.

2. Indicate the capabilities and limitations of ISDN from the point of view of your organization.

Answer:

In my organization there are "islands" of implementation of ISDN technology. ISDN is now viewed as a technology that is appropriate for remote users who dial into their companies' LANs and for some LAN-to-LAN connections. Since my company deals with many external entities (users) this type of network technology is pursued with interest. ISDN can also handle fax traffic. ISDN provides an economical way to establish digital links with remote offices until traffic requirements call for more expensive dedicated lines. The ISDN interface will automatically switch among different devices attached to it, such as a bridge, phone, or fax machine. ISDN also provides a link into the access point of Frame Relay and other fast packet public networks. For example, an ISDN primary rate interface is required to access the AT&T Accunet Switched 1536 Service that lets users dial into 1.536 Mbits/sec lines on AT&T's digital switched network.

Standards for ISDN are being defined by CCITT. The components of the ISDN standard are:

• Support of voice and nonvoice applications using a limited set of standardized facilities

• Support for switched and nonswitched applications

• Reliance on 64-Kbps connections

• Intelligence in the network

• Layered protocol architecture

• Variety of configurations.

The principal benefits to the users in my company can be expressed in terms of cost savings and flexibility. The integration of voice and a variety of data on a single transport system means that the user does not have to buy multiple services to meet multiple needs. The efficiencies and economies of scale of an integrated network allows these services to be offered at lower cost than if they were provided separately. Further, the user needs to bear the expense of just a single access line to these multiple services.

The requirements of various users can differ greatly in a number of ways: for example, information volume, traffic pattern, response time, and interface types. The ISDN will allow the user to tailor the service purchased to actual needs to a degree not possible at present.

The ISDN provides for my company a variety of services, supporting existing voice and data applications as well as providing for applications now being developed. The most important ones are:

• Facsimile

• Teletext

• Videotext.

3. Briefly delineate the major features and projected applications in your workplace of the following BISDN supported interactive services:

a. Conversational services

b. Messaging services

c. Retrieval services

Answer:

In my workplace we are using at some extent all of the above three B-ISDN supported interactive services: Conversational, Messaging and Retrieval. The last two, the Messaging and Retrieval services are already in extensive use since we all use e-mail, and voice mail. Even more there are plans to move toward a distributed file systems that will integrated over the networks and with other than data (text) components, i.e., voice, video, and images.

The Conversational service is not quite up-to-speed, so to speak. We are using teleconferencing but the use of video conferencing is very limited. However, experiments are being made to connect various business partners (remotely located) to our company and exchange various types of information: text, video, voice, graphics. In one of these experiments we used a PC- or Macintosh-based system called CUCME that allows a basic video and tele-conferencing including the use of white board. Even though use a B-ISDN environment most the experienced problems were related to time of day (i.e., network traffic).

4. How does BISDN differ from narrow band ISDN?

Answer:

The original ISDN standard is a narrowband ISDN. The merging standard called broadband ISDN (B-ISDN) operates in the megabit-to-gigabit range. Narrowband ISDN has a maximum rate of 2 Mbits/sec and works over copper cable. Narrowband ISDN can provide 64- or 128 Kbits/sec of digital service over wide-area links, depending on how it is configured. Compare this to typical modem rates of 28 Kbits/sec or less. ISDN provides a significant improvement over existing modem-connected lines, but it still doesn't compare to the 10 Mbits/sec rates of an Ethernet local area network (LAN). However, it's also important to realize that Ethernet LANs are shared by multiple users, so the data rate is 10 Mbits/sec only when no other users are contending for the LAN.

In 1988, as part of its I-series of recommendations on ISDN, CCITT, issued the first two recommendations relating to broadband ISDN (B-ISDN). CCITT modestly defines B-ISDN as " a service requiring transmission channels capable of supporting rates greater than the primary rate." With B-ISDN services, especially video services, requiring data rates orders of magnitude beyond those that can be delivered by ISDN will become available.

The B-ISDN differs from a narrowband ISDN in a number of ways. To meet the requirements for high-resolution video, an upper channel rate of on the order of 150 Mbps is needed. To simultaneously support one or more interactive services and distributive services, a total subscriber line rate of about 600 Mbps is needed. In terms of today's installed telephone plant, this is stupendous data rate to sustain. The only appropriate technology for widespread support of such data rates is optical fiber. Hence, the introduction of B-ISDN depends on the pace of introduction of fiber subscriber loops.

Internal to the network, there is the issue of the switching technique to be used. The switching facility parameters (e.g., burstiness). Despite the increasing power of digital circuit-switching hardware and the increasing use of optical fiber trunking, it is difficult to handle the large and diverse requirements of B-ISDN with circuit-switching technology. For this reason, there is increasing interest in some type of fast packet-switching as the basic switching technique for B-ISDN. This from of switching readily supports anew user-network interface protocol known as asynchronous transfer mode (ATM).

Task 4

1. Discuss the key characteristics of the ATM cell.

Answer:

ATM (Asynchronous Transfer Mode) is a data transmission technology that has the potential to revolutionize the way computer networks are built. Viable for both local and wide area networks, this technology provides high-speed data transmission rates and supports many types of traffic including voice, facsimile, real-time video, CD-quality audio, and imaging.

The ATM makes use of fixed-size cells, consisting of a 5-octet header and a 48-octet information field. In the 1988 specification document, the header fields and header size are undefined, and it had not yet been decided to use fixed-size cells. The definitive decision to use fixed-size cells is documented in the 1990 Recommendations.

There are several advantages to the use of small, fixed-size cells. First, the use of small cells may reduce queuing delay for a high priority cell, since it waits less if it arrives slightly behind a lower-priority cell that has gained access to a resource (e.g., the transmitter). Secondly, it appears that fixed-size cells can be switched more efficiently, which is important for the very high data rates of ATM.

The ATM cell header format depends on if it's at the user-network interface or at the network node interface. The user-network interface cell header contains the following fields: generic flow control, virtual path identifier, virtual channel identifier, payload type, reserved, cell loss priority, and header error control. The network node interface cell header contains: virtual path identifier, virtual channel identifier, payload type, reserved, cell loss priority, and header error control.

The Virtual Path Identifier (VPI) constitute a routing field for the network. It is 8 bits at the user-network interface and 12 bits at the network-network interface, allowing for more virtual paths to be supported within the network.

The Virtual Channel Identifier (VCI) is used for routing to and from the end user. Thus, it functions much as a service access point.

The Payload Type field indicates the type of information in the information field. A value of 00 indicates user information; that is, information from the next higher layer. Other values are for further study.

The Cell Loss Priority is used to provide guidance to the network in the event of congestion. A value of 0 indicates a cell of relatively higher priority, which should not be discarded unless no other alternative is available. A value of 1 indicates that this sell is subject to discard within the network.

Each ATM cell includes an 8-bit header error-control field that is calculated based on the remaining 32 bits of the header.

2. With the aid of an example, describe how voice and data traffic is segmented into cells and switched through an ATM network.

Answer:

I'll use the analogy of vehicles on a bridge to illustrate how ATM works and why it is so efficient. Let's think of the bridge as the ATM connection between two remote LANs. If every vehicle were exactly the same size, as ATM cells are, they could be spaced equally apart in traffic and driven at the same speed across the bridge. Consequently, you could accurately predict when a vehicle would reach the other side of the bridge. In real life, however, vehicles come in many different sizes, making traffic predictions difficult. In data communication, variable-sized data packets that provide uncertain delays are not suitable for video and voice applications (unless prioritization methods are used).

Taking this analogy further, suppose you want to transport a busload of people across the bridge. Busses aren't allowed, so groups of four people climb into cars, traverse the bridge, then climb back into another bus on the other side. Likewise, in ATM, packets of data from upper-level applications might need to be split apart, inserted into a number of ATM cells, transmitted, then recombined at the other side of the ATM transmission.

If several busses arrive at the same time, all can begin traversing the bridge simultaneously, it's not necessary to unload one bus first, then the next. Like the ATM cells, cars holding passengers from each bus are allowed to traverse the bridge one after the other. In communications, this technique is used in multiplexing; in ATM, it is used to transfer data from multiple connections simultaneously.

The fixed cell size and multiplexing provide bandwidth on demand to devices that need it. LAN traffic is often "bursty" due to file transfers or other activities that cause peaks in activity. An ATM switch can detect burst in traffic and dynamically allocate more cells to handle a burst from a particular source.

3. Briefly describe the purpose and functions of the ATM Layer and ATM Adaptation Layer.

Answer:

ATM was originally defined as part of the B-ISDN, which was developed in 1988 by CCITT, as an extensive of ISDN. The B-ISDN and therefore ATM, has three layers:

• The Physical Layer defines the electrical or physical interface, line speeds, and other physical characteristics.

• The ATM Layer defines the cell format.

• The ATM Adaptation Layer defines the process of converting information from upper layers into ATM cells.

The ATM Layer defines the structure of the ATM cell. It also defines virtual channel and path routing, as well as error control. ATM cells are packets of information that contain a "payload" (data) and header information that contain channel and path information to direct the cell to its destination.

The cell is 53 bytes in length, 48 bytes of which is the payload and 5r bytes of which is the header information. Note that header information is almost ten percent of the cell, which adds up to extensive overhead on long transmissions, as pointed out by some ATM detractors. They advocate variable-length technologies like Frame Relay for this reason. The information held in the header is the following:

1. Generic Flow Control (GFC). This field is still being defined, but ATM Forum has defined it to provide a way fro multiple workstations to use the same User Network Interface (UNI). Other possibilities include a definition for the type of service.

2. Virtual Path Identifier (VPI). Identifies virtual paths between users, or users and networks.

3. Virtual Channel Identifier (VCI). Identifies virtual channels between users, or users and networks.

4. Payload Type Indicator (PTI). Indicates the type of information on the payload area, such as user, network, or management information.

5. Cell Loss Priority (CLP). Defines how to drop certain cells if network congestion occurs. The field holds priority values, with 0 indicating that a cell cannot be dropped.

ATM Adaptation Layer (AL) converges packets from upper layers into ATM cells. Recall that each cell has a 48-byte "payload bay." The AAL would segment a 1,000-byte packet into 21 fragments and place each fragment into a cell for transport. The layer is divided into two sublayers. The convergence sublayer (CS) accepts the higher layer data and passes it down to the segmentation and reassembly (SAR) sublayer. The SAR is responsible for breaking the data up into 53-byte ATM cells. If cells are arriving, the SAR reassembles the data in the cells and passes it to upper layers. There are several AAL types:

1. Type 1 is an isochronous, constant bit-rate service for audio and video applications. It is similar to T1 or T3 and provides a variety of data rates.

2. Type 2 is an isochronous variable bit-rate application like compressed video. The carriers have not implemented this interface.

3. Type 3/4 supports bursty LAN-type variable bit-rate data that supports frame relay and SMDS interfaces.

4. Type 5 supports a subset of Type 3/4 functions, providing message mode and nonassured operation. This mode will most likely be quickly deployed.

4. Highlight the distinctive features of ATM switches.

Answer:

An ATM network contains ATM switches, which are generally multiport devices that perform cell switching. When a cell arrives at one port, the ATM switch looks at the cell's destination information and sends it to an appropriate output port. There are fabric-type ATM switch designs that have many ports and are used by the public carriers, and there are bus-based switches with fewer ports that are more suitable for LANs. If multiple ATM switches are connected together, a routing protocol is needed so the switches can exchange look-up connection tables.

One reason for the high switching speeds in ATM switches is that they perform their switching operations in hardware. ATM switches avoid the Network layer (relative to OSI protocol). Instead, ATM simply places information into cells and sends it off. ATM is a so-called "fast-packet" technology like frame relay and Switched Multimegabit Data Service (SMDS), in that it performs no error checking and is not bogged down by such matters. A receiving station is responsible for making sure it received everything from a sender. If a cell is lost or corrupted, the end station must request another from the sender. ATM is not responsible for recovering the cell. In contrast, X.25 uses extensive error checking as packets traverse the network. A packet must be received and error-checked at a node before it is forwarded. This overhead limits throughput. X.25 is designed for older analog phone systems that were prone to error. The error checking was designed to detect corrupted packets as soon as possible. ATM assumes the use of high-quality, error-free transmission facilities.

5. Explain the role of the collapsed ATM backbone in facilitating LAN interconnection.

Answer:

A "backbone" is a network that connects two or more local area network (LAN) segments or subnetworks and provides a data path for packets transmitting information among them. A bridge or router connects each network segment to the backbone. Fiber-optic cable systems such as Fiber Distributed Data Interface (FDDI) are often used for the backbone connection. The backbone cable can extend throughout the premises. Alternatively, a hub device that serves as a "collapsed backbone" can provide a single connection point for all the subnetworks. Backbones handle internetwork traffic while local traffic is handled by subnetworks. Servers attached directly to the backbone provide better access for internetwork users than if they are attached to subnetworks.

Also, a "collapsed backbone" is a backbone that is reduced to fit within a single box. Instead of deploying the backbone cable throughout the enterprise, a cable is attached to the hub from each subnetwork. The trend is natural. Once you've moved servers and other types of network equipment to a central location for management purposes, it makes sense to simply connect them to a single box, called a hub, that replaces the backbone cable. Hubs are built on a chassis with high-speed, often proprietary, backplanes that provide a high-speed bus for expansion boards. Hubs are modular and expandable. Many accept Ethernet, token ring, FDDI, and WAN modules, as well as diagnostic and management modules.

Structured backbone systems make distributed computing easier to implement, configure, secure, and manage. While "collapsed backbone" hub devices present a single point of failure that can cripple the entire enterprise network, the collapsed backbone-in-a-box is still appealing to managers, and it has its place. Some vendors see its benefits and are already designing "super-hubs" built on ATM that include bridges, routers, management modules, and high-speed ATM switching. ATM cell switching connects the bridges and the routers, and in more advanced units, provides connections for workstations and other LAN segment nodes as well.

Task 5

Write a 9 to 10 page describing strategies and guidelines for developing a new networking application for your organization. Examples of applications include desktop videoconferencing, distance learning, distributed corporate training, a traditional for its selection, an assessment of advantages and limitations, and guidelines for implementation. A brief list of references should be included as well.

Answer:

The following are the requirements, strategies, guidelines and plan for development of a new networking environment at my workplace.

Network Problems

Existing LAN infrastructure, called ILAN (Institutional LAN) is a low speed CATV backbone installed over 10 years ago. Over 200 local user-operated networks provide network connectivity from the backbone to the workstations. ILAN low speed CATV RF cable system will be decommissioned for network data by end of FY ‘96.

The corporate computer network environment has been limited by funds and policy to the operation and maintenance of the ILAN backbone networks. The delivery of network services to the workstation via sub-networks has been left to the individual user organizations. Each organization installed sub-networks without regard to the needs of other organizations in the same facility. The sub-networks were installed with varying degrees of quality, capacity, architectures and standards. Re-location of workers in this patchwork network environment results in high costs.

Network Vision

The network vision is defined around the following attributes:

1. Multi-Gigabit digital network

2. Network installed in every office, company-wide, conference room, etc.

3. Bandwidth tailored to user needs

4. Standard user interface to enable changes in user location or user platform at no cost

5. Integrated data, voice, and video services

6. Institutionally provided network administration

7. Robust architecture to accommodate and take advantage of unforeseen changes in technology

8. Be affordable.

Corporate Network Environment Goal

Provide reliable and robust voice, data and video communications which meet the changing needs of customers and enable the easy exchange of information internally and externally. Provide required communication capabilities to every work location, company-wide.

Proposed Architecture

The proposed architecture is called Hi-speed NETwork (HiNet). The following will be the top level capabilities of HiNet:

1. This will be a Company-wide High Speed Digital Network, starting as soon as approvals will be in place.

2. HiNet will be replacing the ILAN low speed connections with connections to the ILAN - Fiber Optic System.

3. HiNet will provide for a permanent network connection infrastructure for easy, low-cost workstation setup, moves and changes.

4. HiNet will consolidate network operations into a centralized team that will provide 24 X 7 coverage, Preventive maintenance of the network, Performance and Network management, and Configuration management.

5. HiNet will make use of the commercially-available hardware and software, be robust and reliable, allow for flexible configurations for low-cost additions/moves/changes and provide for network growth.

The major milestones of this projects are (tasks):

A. Complete the ILAN (Institutional LAN) Fiber Optic System by the end of FY ‘96:

a. Provide FDDI 100 Mbps to 44 buildings

b. Extend fiber-optic links to all buildings requiring it

c. Move toward an ATM based backbone in FY ‘97.

B. Provide full network capability to the wall for every office and workstation by end of FY ‘97:

a. Wiring of Fiber Optic cable capable of 155 Mbps

b. Hub architecture at 10 Mbps upgradeable to 155 Mbps

c. 155 Mbps delivery to the workstation by FY ‘98.

C. Provide Institutional installation, operation, maintenance and management for the entire network.

D. HiNet will set the foundation for CyberNet 2000 by paving the following capabilities:

a. ATM could be used with the FDDI ring

b. Use of Synchronous Optical Network (SONET) to 10 Gbps

c. Use of ATM switches.

Clarifications/Definitions

HiNet is an effort to provide a high reliability, easy maintainability, and high speed capable network infrastructure. This new infrastructure will consist of a structured cabling system which will support voice, data, and video communications at up to 15 times the current network speeds. It will be institutionally supported 24 hours a day, 7 days a week and all corporate provided network equipment will be monitored centrally. Institutional monitoring will provide proactive bandwidth management and problem identification and resolution. This project will be undertaken by the Network Engineering Department.

Current ILAN Configuration

ILAN backbone is currently configured as follows:

Five 5 Mbps Channel Broadband Networks

4 A/R Ungerman-Bass Net/One XBS

5/S Scientific and Engineering Backbone TCP/IP

3/P Multiprotocol backbone AppleTalk DECnet, IP, IPX, XNS

6/T DECnet

4/Q Unused.

Alternatives Considered

The following were design alternatives considered:

Shared Ethernet (10BaseT, 10 Mbps, CSMA/CD)

Switched Ethernet (10BaseT, 10 Mbps)

Full Duplex Ethernet (10BaseT, 20 Mbps)

100Base-Tx (100 Mbps, CSMA/CS)

100VG-AnyLAN (100 Mbps)

FDDI/CDDI (100 Mbps)

Alternatives Not Currently Considered

The following were design alternatives not considered:

ATM

Wireless

General Networking Information

The following is a general information area for those who are not familiar with the network concepts and technology.

1. Cabling

The first part of the HiNet project involves designing a new cable system which will be able to support the high speed specifications as they become finalized and supported by products in industry. The network can only be as good and as reliable as the underlying cabling structure. This new structured cabling system should support most of our current investment in desktop hardware and allow users to migrate to higher speeds as needed. HiNet will deal with 3 discreet cabling components:

a. The completion of the fiber optic backbone (ILAN_FOS) which ties the major buildings together.

b. The installation of fiber optic riser cabling to tie the floors of a building to the main connection point to the ILAN_FOS.

c. The installation of horizontal wiring on each floor of a building to provide connectivity to each work area.

2. Transceivers

The following type of transceivers will be employed as part of the proposed architecture:

Sonic Apple AUI Ethernet Transceivers

Farallon Apple AUI Ethernet Transceivers

BNC Connector Description

Digi Fast Ethernet Products

Digi Fast Ethernet Transceivers

3. Mini-Huns

The following mini-hub will be employed as part of the proposed architecture:

Sonic Mini Hub.

4. Network Interface Cards

The following network interface cards will be employed as part of the proposed architecture:

Intel fast Ethernet White Paper

Intel Pro100

Farallon Fast EtherTX-10/1000 PCI and NuBus Cards

Sun Fast Ethernet Adapters.

5. Signaling Specifications

Various specifications govern how data is transferred at the physical layer (the cable). The following is a list and descriptions of those specifications that will be talked about in HiNet. Please keep in mind that the cabling system installed by HiNet should support all of the high speed signaling specifications listed below. Only electronics on either and (workstation and Hub) must change to support faster signaling rates.

a. Ethernet

Much of the current cabling within the corporation is Ethernet of both the Thin and Thick variety. The following sources can be further consulted fro specific information:

Introduction to Ethernet and the OSI Model

Ethernet Topology

Ethernet: Access to Ethernet/IEEE 802.3 Information

b. 10BaseT

Recently installed cabling within the corporation has been 10BaseT. Couple of buildings can be used as examples of 10BaseT installations. The following sources can be further consulted fro specific information:

Guide to 10 Mbps Ethernet

c. Switched Ethernet

This extension of Ethernet can be implemented with virtually no change to a user’s workstation and givens a user access to a dedicated 10 Mbps of bandwidth rather than the traditional shared 10 Mbps of bandwidth used in 10Base-T and the coaxial technologies (Thick and Thin Ethernet).

d. Fast Ethernet

Fast Ethernet is another type of technology that will be considered as part of the overall design of HiNet.

e. Asynchronous Transfer Mode (ATM) and Gigabit Communications

ATM seems to be the networking wave of the future. Primarily because it will allow video and other sequentially dependent data to be delivered at high speed. The Currently the corporation has an ATM testbed. The HiNet design mandates that the cabling system to be installed to support ATM speeds (155 Mbps).

f. Network Management and Performance

Another component of the HiNet design will be the Network Performance Management capabilities, as follows:

Enhanced Network Management Platform:

Standards based

Network Diagrams of the HiNet Infrastructure (logical network, physical network, available on-line to designed users)

Notification of Network Component Failure (central management station, local network/system administration).

g. Protocol

The chosen protocol for HiNet is TCP/IP.

Recommendations

Management recommendations are:

1. This project should be number 1 priority for the network engineering department

2. Tailoring requirements to current occupants as fast as possible

3. A Task Manager and Team will be appointed to organize the various elements involved and assign pertaining sub-tasks

4. Invite participants and open to anyone that is interested

5. Formal design reviews should be held.

Proposed Schedule

The following is the proposed schedule for the implementation of HiNet:

1. FY ‘96 tasks:

Requirements Gathering

System Engineering

Testbed Installation

Establish Cable System Subcontract

Initial Hub Procurement

Begin Installation

Cable Installation

Hubs

Conversion of Existing Building Networks

Establish Configuration Management System

Establish Network management System

2. FY ‘97

Complete Installation

Complete Conversion of Existing Networks

Deliver Operations

Establish maintenance.

References

Intel Corporation. (1996). Fast Ethernet Overview and Deployment Guide., Palo Alto, CA.

McNamara, John E. (1985). Local Area Networks, An introduction to the technology. Digital Press.

Digital Equipment Corporation. (1990). Network Troubleshouting Guide. Digital Press.

IEEE. (1990). Introduction to Ethernet and OSI Model. Piscataway, NJ

Black Box Corporation. (1994). Ethernet Topology.