|
|
|
| P is for Practical | |
| PerlMonks |
Re^3: Bidirectional lookup algorithm? (Updated: further info.)by flexvault (Monsignor) |
| on Jan 06, 2015 at 11:10 UTC ( [id://1112320]=note: print w/replies, xml ) | Need Help?? |
|
When programming my database engine, I found that 'index' did best with smaller scalars ( reason for mentioning 'database sharding' ). My times were close to the worst case ( worst case is 'not found' ). I found that with a 8-byte key, the 'index' technique I used was 4.7 times faster searching 512 byte records than searching 1,024 byte records. Naturally, 'index' searching a 5MM+ byte record is going to be real slow! My attempt was to give you another approach, not a final solution. Good Luck...Ed "Well done is better than well said." - Benjamin Franklin
In Section
Seekers of Perl Wisdom
|
|
||||||||||||||||||||||