Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
64 Cards in this Set
- Front
- Back
|
What are buses? |
A collection of (almost always copper) conductors, each transmitting a different signal between different parts of the computer. |
|
|
What do parallel buses have? (3) |
Address lines, data lines & control lines. |
|
|
What happens on each bus cycle? |
The CPU asserts whether a read or write is to take place & sets the address lines. The transmitter sets the data lines. The receiver reads them & copies the data to registers or memory locations. |
|
|
What does the bus master do? |
It asserts on the control lines asking all other devices to be its slaves & then it controls the transfer of information until it gives up control. |
|
|
What is bus arbitration? |
When multiple devices try to assert as the bus master, bus arbitration is how it gets resolved. |
|
|
What is multiplexing? |
Using the same lines for data and addressing. The address must be set first, then the data. |
|
|
How many lines does a PCI bus have? |
124 or 188, with multiplexed address lines - 32 or 64 lines for addressing & 16 control lines. |
|
|
What is the data transfer rate for a 66MHz PCI bus?
|
528Mbytes/s. |
|
|
What does synchronous mean? |
Everything works to the clock & at the speed of the slowest connected device. |
|
|
What are used for 3D graphics and other speed-requiring applications? |
PCI Express (PCIe).
|
|
|
How does PCIe work? |
It works more like a network - it has a central hub which all devices connect to. Data is transmitted/received simultaneously through lanes & devices can use multiple lanes to get speeds of Gbytes/s. |
|
|
What are the 4 wires in USB 2.0 used for? |
2 for power, 2 for data. |
|
|
How does USB transfer data? |
In packets, which are wrapped up in frames - 1 bit at a time. |
|
|
How fast is USB 2.0? |
It has a data transfer rate of 60Mbytes/s. |
|
|
How is USB 2.0 connected to the main computer bus? |
By a root hub. |
|
|
Why are interrupts necessary? |
Devices may need attention from the CPU at unpredictable times (e.g. when a mouse is moved/clicked). |
|
|
How are interrupts handled? |
Using the bus - The CPU saves the program counter and jumps to a device specific interrupt handling subroutine (part of the kernel). |
|
|
How can a network be classified? (2) |
By its scale (LAN/MAN/WAN). By its topology (Bus/Star). |
|
|
How do nodes (i.e. computers, etc.) connect to the network? |
By using network interfaces or network interface cards. |
|
|
What does a repeater (or hub) do? |
It regenerates any signal passing through it, sending the data packets to every node connected to it, no matter of the packet's destination. |
|
|
What does a bridge do? |
It examines each packet and only lets trhough the ones whose destination is on the 'other side' of it. |
|
|
What does a switch do? |
It has multiple nodes connected to it, and only forwards packets down the node correct for its destination. |
|
|
What does a router do? |
It connects different networks (such as a LAN or a WAN), translates network protocol addresses between the networks & filters out packets. |
|
|
What is the Internet formed of? |
A freely-growing collection of networks that can communicate with one-another. |
|
|
What is the Open Systems Connection Basic Reference (OSI) Model? |
An example of abstraction to provide a virtual network that hides details from the programmer and the user. |
|
|
What is the physical layer concerned with? |
The physical process of getting bits from one location to another (voltages in wires, wireless frequencies, etc.). |
|
|
What is the data link layer concerned with? |
Error correction, grouping bits into frames & ensuring frames make the next hop of their journey correctly. |
|
|
What does the network layer rely on? |
A network address. |
|
|
What is the network layer concerned with? |
Ensuring the full journey of a packet through the network is successful - if a direct connection between its source & destination is not available, it is responsible for routing the packets through computers in the network. |
|
|
What is the transport layer concerned with? |
Wrapping up the data into datagrams (packets) ready for transport by the network layer, flow control, putting packets in the right order & recovering from transmission failures. |
|
|
What is the session layer concerned with? |
Managing dialogues between machines that are exchanging information. |
|
|
What is the presentation layer concerned with? |
Translating the data into a form that the session layer can handle. |
|
|
What is the application layer concerned with? |
Acting on requests from the user's program. |
|
|
How many nodes are possible with 32 bit addresses? |
4.3 billion. |
|
|
What is dotted decimal notation? |
A form used for writing an internet address of a node - e.g. 139.184.24.11. |
|
|
What happens when a request is made using an alphabetical address? |
DNS lookup is performed - a request is sent to a name server which is constantly updated & has the dotted decimal address relating to the alphabetical address requested. |
|
|
Why is DNS lookup good? (2) |
Servers/computers can be changed whilst the web address remains the same.
A single computer can be the location of multiple web addresses. |
|
|
Where does Internet Protocol operate, what uses it & where is it supported? |
It operates in the internet layer, is used by the transport layer & is supported by the link layer. |
|
|
What does it mean to be a connectionless protocol? |
It means it launches packets towards the destination without a prior connection being established. |
|
|
What does it mean to be an unreliable protocol? |
It means it does not report errors to the sender or recipient - packets may arrive in any order & be lost, duplicated or corrupted. |
|
|
Where does Transmission Control Protocol operate, what uses it & where is it supported? |
It operates in the transport layer, is used by the application layer and is supported by the internet layer. |
|
|
What does TCP do? |
It receives data in the form provided by the application and splits it into packets, ready to be given to IP. It reassembles packets received over the internet layer into the correct order for the application. It carries out flow control to slow down transmission if the receiver can't keep up. |
|
|
What does it mean to say a protocol is reliable? |
It means it checks that all the packets arrive & are correct - it will ask for retransmission if necessary. |
|
|
What does it mean to say a protocol uses connections? |
It means it gets acknowledgement from the recipient before sending. |
|
|
What does TCP use to differentiate between packets for different applications at the same destination? |
Port numbers. |
|
|
Where does Hypertext Transfer Protocol operate & what does it rely on? |
It operates in the application layer and relies on the transport layer. |
|
|
What does HTTP do? |
It manages interactions between clients (e.g. a web browser) and a server. |
|
|
What is an example of a scheme name for a URL? |
http:// |
|
|
What is an example of a host name for a URL? |
www.gnu.org |
|
|
What is an example of a path name for a URL? |
/licenses/licenses.html |
|
|
How is the host name used? |
It is used by clients to establish a connection to the server. |
|
|
How are the scheme and path names used? |
They are used by clients to send requests (e.g. GET /licenses/licenses.html HTTP/1.1). |
|
|
How does the server respond to a request? |
It responds with a message that includes a representation of the resource requested (often a text document containing hypertext markup). |
|
|
What is Hypertext Markup Language used for? |
Representing documents so a web browser can effectively display them to a user & allow a user to navigate between related documents. |
|
|
How are items on the document (e.g. images, text, tables, etc.) distinguished between for display? |
Tags. |
|
|
How do search engines work? |
By running a crawler program, building indexes to the found web pages & accepting HTTP(S) requests from users looking for a specific word or phrase, returning the results ordered from most relevant to least. |
|
|
How does the crawler program work? |
By generating HTTP requests & analysing the returned web pages to extract URLs (which generate more requests). |
|
|
What is the Resource Description Framework? |
A formal notation for the specification of semantics - uses subject-predicate-object triples to define the meaning of an element. |
|
|
What does the Web Ontology Language present? |
A family of formal specification languages for the building of ontologies, which enable reasoning over content. |
|
|
What is RDFa? |
A language that extends HTML and other markup languages with additional semantic attributes, which can be directly embedded into the source. |
|
|
What does cloud computing allow? |
The sharing of computer (software/data/processing) resources over the internet. |
|
|
What does Amazon Web Services have? |
EC2 (Elastic Computing Cloud) - virtual private servers. S3 (Simple Storage Service) - remote data storage. Cloudfront - distributing data to locations near users. Elastic Beanstalk - automatic management of applications. |
|
|
What are the advantages of cloud computing? (4) |
Convenience (services don't have to be installed). Rapid scalability. Improved reliability. Costs of set-up become costs of operation (more spread out). |
|
|
What are the disadvantages of cloud computing? (2) |
Reduced control over technical trade-offs. Greater security risk to data stored online. |
