Help (1 Viewer)

[crammy90]

www.google.com
google is the DNS
is a DNS the webserver? or is it part of a webserver :S
"what best describes a distributed database"
b) dB that has different data physically stored on different computers, however this is not necessarily apparent when users access the database
c) the complete database is stored at different locations, each copy being synchronised on a regular basis with all the other copies"

c is the answer
b = a fragmented database
c = a DDBMS which uses the process of replication

would any1 else argue both to be correct?
im sick of these contradictory definitions and how we must just acccept what BOS says they believe!!

 

[ondroo]

DNS = dynamic name system. Its just a general term to describe the conversion from a domain name (www.google.com) to the servers hosting the site's IP address (in google's case 66.249.89.99 which u can type straight into the address bar and it will load google anyway). The whole purpose is to use words instead of unmemorable ip addresses.

So www.google.com is the domain name...

A DNS server is the computer(s) that perform the conversion. Each ISP has their own. Also there are some independent ones (e.g. www.opendns.com).

As for that question, you're right in saying they should be both acceptable. Where is it from?

 

[Rampager]

We do the exact same subjects Ondroo [except I do 3-unit math, shame on you], high five buddy.

I would've said C in that multiple choice to be honest, but maybe this is because I learned a different definition of distributed database to you, Macquire IPT Revision guide says this:

"A distributed database is located in more than one area. The purpose of distributed databases is to lessen demands on the overall database if many requests for information are coming from remote sites... Where a distributed database requires users to update information as well as be able to access it, care must b etaken to ensure that if data is mirrored on more than one site, all sites are regularly updated."

 

[crammy90]

where from?

it was from a 2005 trial exam

We do the exact same subjects Ondroo [except I do 3-unit math, shame on you], high five buddy.

I would've said C in that multiple choice to be honest, but maybe this is because I learned a different definition of distributed database to you, Macquire IPT Revision guide says this:

"A distributed database is located in more than one area. The purpose of distributed databases is to lessen demands on the overall database if many requests for information are coming from remote sites... Where a distributed database requires users to update information as well as be able to access it, care must b etaken to ensure that if data is mirrored on more than one site, all sites are regularly updated."

yeh thats the definition i learn all year and then i was reading from sam davis and got confused. I dno if i should say it as it may confuse you also lol. Ill do it anyway:
"A distributed database is a set of connected databases stored on multiple computers that appears to users as a single dB" - definition
Ill just summarise the types:

• Fragmented: this is B: different parts of the database are stored at different locations (so each instance of a record is only stored once).
  EXAMPLE: just say you have "Brads car repairs" and theres 4 franchises in sydney. A customer comes in and they want their car repaired. As the DDB is fragmented, when searching to see if he is an existing customer (in the Customers table), the "Customers table" at every franchise needs to be searched (as the customer could have had repairs at another location prior). So collectively, the SQL is carried out and it appears as if it is searching just 1 database, when really it is searching all Customer tables and different sites.
• Downloading: not listed in question. This is basically like fragmented where there are different parts of the data stored physically at different sites. What happens, in the case of Brads car repairs, after this SQL for the new customer is performed (from above), if a record is found indicating the customer has had repairs at another franchise, then this record is "downloaded" from the remote database and this record it stored at this site where the customer is now getting repairs. (doesnt mention anything about data redundancy but this seems really stupid as then youd have 2 records for the same person and so when u did queries you could get conflicting results if the customer changes some of their details at only 1 site :S eh i dno)
• Replicated DDB: this is C: its where each dB has exactly the same data stored in their database. They are all synchronised so to speak. That happens basically is 1 dB is made the Master, the others are "Replicants". I think of the master like a hub. The Replicants send their data to the hub, which is then sent to all databases so they each recieve the same information; only different between this and a hub situation is that all replicants accept the signal. So basically all dB get the same info as eachother be sharing through the master. Any updates in the master are copied to all replicants also.

So now you can see why i was confused sami lol. Im starting to think with SDD and IPT that all these conflicting resources and stupid syllabuses are the only things making these courses hard lol. And how even if something isnt right, if thats their answer you have to learn it

 

[Rampager]

Hahaha, yea, now I see the problem. That's weird, Sam Davis knows too much, makes shit complicated I've never heard about the different types of distributed databases, basically only the 'Replicated DB' from my textbook so *shrug*.

Oh well, it's just a multiple choice question aye, only worth a mark, no need to get antsy Yea the SDD and IPT courses make up their own shit, and even if you know the proper real-life term it all goes out the window but what are you going to do?

 

[crammy90]

Hahaha, yea, now I see the problem. That's weird, Sam Davis knows too much, makes shit complicated I've never heard about the different types of distributed databases, basically only the 'Replicated DB' from my textbook so *shrug*.

yeh i used this multiple choice as an example because distributed databases was needed in a scenario question worth 5 marks and i couldnt be bothered typing it aha so i just looked for a multiple choice. and then even the multiple choice i didnt understand. So my plan of attach was to use whatever info i gathered from this multiple choice to use for the scenario

 

[ondroo]

We do the exact same subjects Ondroo [except I do 3-unit math, shame on you], high five buddy.

I was doing 3 unit maths for beginning of prelim but u see i'm laazzzyy and couldn't be bothered doing all the homework haha. Plus I need time to properly understand stuff so it actually absorbs. Lol im beating nearly the whole 3 unit class in 2 unit (ranked 2nd)

i was reading from sam davis and got confused.

I think the davis explanation is the best one. Just look at the diagrams, they make sense to me. Fragmented (different parts of table in different locations), downloading (downloads a copy of database from different locations), replication (synchronises - all databases hold same information at same time)

Im starting to think with SDD and IPT that all these conflicting resources and stupid syllabuses are the only things making these courses hard lol

Yeah thats one thing I hate about the IT industry - its so new so definitions aren't properly established. Like in SDD there are like 5 different definitions for incremental compilation! And in the real world all the compilation methods are rarely separated like they are in the syllabus cause some languages combine them.

 

[crammy90]

Yeah thats one thing I hate about the IT industry - its so new so definitions aren't properly established. Like in SDD there are like 5 different definitions for incremental compilation! And in the real world all the compilation methods are rarely separated like they are in the syllabus cause some languages combine them.

yeh whats ur def ur writing for incremental comp
a) the "its when you compile the program and then make changes to source code and only those changes get compiled"
or the other
b) i think its something like translates the lines of code and compiles already translated sections or some bs.

 

[ondroo]

depends on the context of the question. But if asked generally i'd prob write both just to cover bases

 

[mijoe]

Btw, DNS is not necessarily Dynamic Name System, not even sure if that term exists. It is used commonly to refer to Domain Name Servers, but can also refer to Domain Name System, which is what ondroo might have been trying to say. His definition was correct, but for an actualy answer, you need to expand much further.