4.4 File Organisation

Question 1

Serial File

Answer 1

Stores records in the order they are recieved

• difficult to locate specific record as whole file has to to be traverse, not efficient for searching

+ ideal for small data sets, simplistic to implement

Question 2

Sequential File

Answer 2

ordered in a specifc mannor like aphabtically

+ more efficient searching as its ordered
- writing data, if you need to insert something in the middle it can be difficult as the whole file needs to be read and rewritten

Question 3

Indexed sequential file

Answer 3

Same as sequential file but each section (not record) has an index which allows a broad search for a section that contains a record, then an more in depth sequential search into that section to find the record.

• index takes up storage space and requires updating
  + extremely fast searching

Question 4

Random access file

Answer 4

inputs data key, then the hashing algorithm uses the key to produce an address of a record. Use the same data key to return that record.

+ instant access, no reading through data

• collisions can occur when hashing algorithm produces the same address for two or more records. They can be resolved using an overflow file (which uses sequqntial storage) which when a record is assigned a taken address then its put in this area. So when locating at this address it returns the record at this address and then the system knows to check the overflow file for the other record. Can also be resolved using linear probing

Question 5

Multi level index

Answer 5

Speeds up access by creating index ‘ranges’ that point to another index, as you follow the index link you get closer to the actual data. The last index cotains the physical address of the record. This index link may continue several times.