Organisation and Structure of Data*

• The purpose of a hashing algorithm is, to convert a string of characters into a fixed-length key using a calculation.
• Random access in files involve jumping to a record directly without the need to search through an index.
• Direct access files are split into blocks where block contains a fixed number of records.
• A size of a block is the product of the number of the records and the size of each byte.
• The load factor is the number of records stored over the maximum number of records.
• Deterministic is a hashing function that ensures that when searching for a key, the same result is always produced.
• Uniformity is a hashing function that ensures all keys are evenly spread over the block range. This is to reduce block overflow.
• Data normalisation is a hashing function that ensures all keys are normalised (in the same character format i.e. all lower case or numbers).
• Continuity is a hashing function that allows keys to differ if they were originally identical.
• Non-invertible hashing function means that it is not true that the hash value can be reversed to get the original key.
• Collisions occur when two different keys produce the same hash values. This causes longer linear searches as records get placed in the same block.
• Using multipliers can reduce collisions to improve distribution.
• Block overflow is caused when collisions are too great and a block has no more available storage for further records.
• To solve block overflow, an overflow area can be created but essentially defeats the purpose of a direct access system. Or to redo it all and create more blocks to cater for the overflow.
• Hashing is used for indexing/retrieving data in a data structure.