Block 6 - Computers yesterday, today and tomorrow

Objective 1 Distinguish between true and false statements concerning the terms in the index for this block, or explain the terms in your own words.

Objective 2 Outline some historical developments in computers and their applications.

• Mid 1940s - first electronic digital computers built; intended for military applications
• Early computers very expensive and needed a lot of maintenance; used by large business/scientific companies
• Mainframe computers = large computers designed to store and process large amounts of data
• Early 1960s - minicomputers introduced at a price that made it possible for them to be used for a single task.
• Today - workstations and desktop computers are used
• Supercomputers = computers that use parallel processing so are very fast; used to forecast the weather (!!!)

Objective 3 State and explain some of the principal differences between DSP chips and general-purpose microprocessors.

DSP (Digital signal processing) chips are a form of dedicated processing chip. Their instruction sets are specially tailored to produce digital signals. They are used in such applicationns as digital mobile telephones.

Some of the principal differences are:

1. The instruction set - instructions are chose to match operations frequently carried out in signal processing, can carry out multiplications in a single processor cycle
2. The registers - there are several accumulator registers to hold the results of multiplications
3. Cache memory - no advantage in having a data cache, programmer can choose which instructions should be present in the cache and therefore keep very close control of the time taken to execute instructions
4. Programming - DSPs are often programmed in assembly language

Objective 4 Sketch and describe configurations commonly used for LANs, and explain the differences between hubs, bridges, switches and routers.

The three commonly used configurations for LANs are bus-based, star and ring. (See page 14 for diagrams).

The differences between the various pieces of network equipment are:

• hub = a connection point with multiple ports for devices on a network. When a packet arrives at one port, it is copied to all the other ports so that all computers connected to the hub can see the message.
• switch = a connection point with multiple ports for devices on a network. When a packet arrives at one port, it is forwarded only to intended destination, not to all other ports on the switch
• bridge = a device that connects two LANs or segments of a LAN. The two LANs being connected can be alike or dissimilar (e.g. could connect Ethernet and Token Ring). Bridges simply forward messages without analysing and re-routing them, so they are faster but less versatile than routers.
• router = a device that connects two LANs or segments of a LAN. Routers are similar to bridges but provide additional functionality, such as the ability to filter messages and forward them to different places based on various criteria.

Objective 5 Describe and compare essential features of the IEEE 802.3 and FDDI standards for LANs.

SAQ2 (a) State one important difference between the IEEE 802.3 and FDDI protocols

The most important difference between the two protocols is that IEEE 802.3 uses carrier-sense multiple access with collision detection whereas FDDI uses a token.This means that the circumstances under which computers are permitted to transmit messages are quite different in the two protocols.

(b) State one other important difference between their protocols

• The different speeds - 10 megabits per second for the plain form of IEEE 802.3, 100 megabits for FDDI
• The maxmimum number of computers that can be interconnected - 100 for bus-based IEEE 802.3; 500 for FDDI
• Transmission of signals - IEEE 802.3 is based on cable so uses electrical signals; FDDI is based on fibre and so uses light to transmit signals

Objective 6 Describe the essential features of ATM.

ATM (Asychronous transfer mode) has been designed to handle real-time messages such as voice and video. In ATM, messages must be divided into packets that are exactly 48 bytes long. Each packet is made upinto a cell by adding 5 bytes to contain error-detection and a means of identifying the message.

ATM uses store-and-forward switching, that is each cell is stored briefly at each route before being sent onwards on its path towards the destination computer.

SAQ3 Explain how it is that an ATM cell does not need to contain the address of the destination computer.

Each cell does not need to include the destination address because the route for the cells in any particular message is set up at the outset and each router along the path is told what this route is. All that is needed, therefore, is for each message to be given a unique identifier which each cell formed from it must use, and for the routers to recognize this identifier.

Objective 7 List and briefly explain the tasks that must be performed by network interface cards.

For the transmission of messages, the following sequence is necessary:

• divide the data into packets
• calcualate the error-detection bits
• form each packet into a frame
• transmit the frame onto the network
• detect a collision, stop transmitting and send out the jamming signal
• wait for a random time before transmitting.

SAQ 4 List the tasks necessary for receipt of messages.

• detect the destination address in the frame
• use the error-detection bits to check for errors
• send a signal indicating a faulty frame if necessary
• extract correctly recieved packets from the frames and store them

Objective 8 Explain what IP addresses and domain names are and how they are used in the internet.

IP addresses are 32-bits long and identify a computer on the Internet. When you log onto the Interne from home, your ISP allocates an IP address dynamically (i.e. you get whatever address is free) so that you can access the Internet.

As IP addresses are not very memorable (e.g. 192.0.0.1) domain names are associated with groups of IP addresses (e.g. open.ac.uk). On the Internet, this means that you can send a message using the domain name, and then a domain name server looks up the IP address and sends on the message so that the message reaches its destination.

Objective 9 Describe the essential features of TCP/IP.

In IP all datagrams (frames) must include the IP address of the recipient, which is a 32-bit number and an indication of the size of the datagram. IP is used in conjunction with TCP to care of the reliable delivery of IP datagrams by adding bits to provide error-detection and bits that number the datagram.

SAQ 5 Suggest how the fact that TCP numbers the datagrams helps to make delivery reliable.

If datagrams are numbered then the destination computer can check that all datagrams have arrived and if the datagrams arrive out of order the destination computer can still reassemble the packets correctly into the message.

Objective 10 Draw on the material provided for Section 4 to outline some important trends in computers, microprocessors and microcontrollers.

?????????????????????