Random access files
These allow access to any part of the file rather than moving through sequentially.
The current model we have used is very much serial. This is because, for most files, this is how they will work. We can access a file in random access mode. This means we can jump to ANY part of the file without having to physically read it.
There are some major issues though. Where do we jump to? We will need to know! A index sequential file will use random access to jump to the correct position in the file once the index has been consulted. Also other files which are based around specific data structures would also use random access. A good example would be a binary tree. I wil not get into how a binary tree is stored in a file but I will say that random access is the only way that can work and still keep the speed increases that a binary tree would allow.