Computing OCR A451 3rd June 2015 *Official Thread* ComputerSystems&ProgrammingWatch this thread
There is typically always a big question regarding reliability on these exams, and the same concept applies throughout them. Although it is common sense, try and study some of the factors that contribute towards a system being reliable and why it is important for it to be as such.
For example, reliability is imperative in bank transaction system as inconsistencies in data handling can lead to invalid data.
Software is not just 'programs'. You need to consider the environment that they run in (the operating system) and how this contributes to the running of software, specific types of software and utilities (such as antivirus and the defragmenter), and also you must consider the licencing of software (and as ever, the pros and cons) - that's essentially proprietary and open source: never seen it get any deeper than that.
Algorithms & Programming
Practice writing an algorithm for an everyday encounter. We've seen ones on vending machines and comparison of dog weights (or something along those lines), so the topic is never anything special; most importantly for this, ensure that you do not use a specific high-level language as the purpose of an algorithm is for it to be interpreted by any programmer.
I only recall seeing a flow-chart question once, but it is very possible that it could come up on this paper.
A good way of formatting your pseudocode is by writing your statements in the upper case, ensuring that you write an END for all of your statements (eg - ENDIF), using indentation, and using the lower case for named values (variables).
The role of the CPU is important but easy to study. Remember that it is an electronic component placed on the motherboard which operates on the fetch-decode-execute cycle. Clock speed is determined by the number of instructions per seconds, and as such uses the unit Hertz.
It's good to know all of the topologies in some depth, but being able to simply graphically represent or name a topology has been favored in most past questions. Benefits and drawbacks are often common sense in this part, so if you're comparing Bus to Star, for example:
- Star is more efficient as each node has an individual line connecting it to its respective hub
- Bus has a single line with branches which can lead to data collisions.
- If one line fails in a Star network, the rest of the network remains unaffected (unless the node offers a shared peripheral/storage functionality, such as a server/printer)
- If the bus-line fails in a bus network, the network will typically fail completely.
- Economically speaking, Star requires more cabling (network cable is cat5, cat5e, or cat 6) and is more expensive - this is because there is an individual line from the hub to each node.
LAN & WAN definitions are easy to remember, but there are some less obvious benefits and features of each. A LAN, whilst allowing files to be shared, is also more suitable for allowing printers to be shared - think about some other things, look at mark schemes for popular and "why didn't I think of that?" ideas.
Finally, DNS was a key question on a recent paper. Remember that a domain name 'points' to a numerical IP address: the name is there for convenience of the user, and the IP address there for the computer to understand and communicate with other networks using protocols.
Standards are there to make the world of technology a better place, essentially.
They are designed to benefit everyone who works with devices from the keyboard to the iOS suite by making them cross-compatible and familiar to all.
That's the concept in my eyes, but there's certain subsidiaries of standards such as De Facto and De Jure (unsure of spelling on the latter, don't count me on it) which you would need knowledge of.
mouse goes in
screen goes out
Well, not that simple: consider accessibility for the disabled and how input/output devices have developed. Still pretty simple.
There is a wide array of types of secondary storage. You should be able to evaluate the benefits and drawbacks of each, and thus able to compare them. For me, it always pays to write more about these because I tend to accidentally repeat myself in different wordings about what is essentially the same factor - but you might not be as dumb as me.
This is something where common sense will often not come into play. There's some charming questions about data types which really do give you easy marks if you know that a series of alphanumerical characters is a string and soforth, but when considering data redundancy, foreign keys - you will really need to know what you're talking about. Ensure that you revise these.
Binary Logic is easy when you've got your head around it. You just need to know the functions of the basic gates on the syllabus - get to know truth tables (easy marks!)
Consider volatility in your revision, the purpose of memory, how in a certain scenario a change to memory yields a certain result (obviously more ram = more instructions stored at a time = higher performance),
Practice the units of computing data, what binary and hexadecimal represent, manipulation of both binary and hexadecimal, and of course the lovely types of data compression and how that interlinks with the way that images and sound are represented.