If every record is of a fixed length, the place to find a record is simply (record length * record number)Indeed, but that still leaves you with the problem of compiling the list of the records you need to retrieve. Only if the "sequence" of records can be directly translated into a record number, is a fixed-length record scheme beneficial. Otherwise you will still need to build some kind of index which translates between the "sequence" and the record numbers.
It is my experience that one leaves such things best to the database server. Any well-built database server should be optimized to make indexing and retrieval efficient and fast. I do not think any perl-based solution can compete with that in any other than the most trivial of cases. Just consider the time needed to save and load an index (hash?) of several "hundreds of millions" of keys and record-numbers. A modest estimation of say 25 bytes per key / value pair, gives you an index on disk (in YAML or JSON or Data::Dumper format) of several GigaBytes, which you will be hard pressed to keep in memory as a hash.
CountZero
A program should be light and agile, its subroutines connected like a string of pearls. The spirit and intent of the program should be retained throughout. There should be neither too little or too much, neither needless loops nor useless variables, neither lack of structure nor overwhelming rigidity." - The Tao of Programming, 4.1 - Geoffrey James
In reply to Re^2: fast+generic interface for range-based indexed retrieval
by CountZero
in thread fast+generic interface for range-based indexed retrieval
by jae_63
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |