Which is the best lossless database compression technique

I have a database with millions of records in it.

the table has two attribute 1] Name 2] Location

there are many records which has different name but s开发者_运维技巧ame location so is there any compression technique that can be applied on the location attribute?

You can suggest the various loss-less compression techniques with their supported database names.

Normalise your database. Have a LOCATIONS table and give your first table a foreign key into it.

Less than 65,535 distinct locations? That's two bytes per record in your original table. Pretty good compression if you ask me. Also makes indexing and comparisons faster.

Data is more important than code. Get the data right first. Normalise as much as possible by default. Denormalise only after you've discovered a (specific, measured, and fully-attributed) performance problem or some great cause.

You want to look for a spatial index. A si reduces the 2d complexity to a 1d complexity. A si is used in heatmaps and treemaps. It looks like a quadtree.