How about the relationship of compression rate, query efficiency and data？

1

How about the relationship of compression rate, query efficiency and data？
For example 10000 records:

  1. 1000 unique values, each value 10 records
  2. 100 unique values, each value 100 records
  3. 10 unique values, each value 1000 records

Which cases can get good compression rate and query efficieny please? Will case 3 get the best compression rate but worst query efficiency as query need to decompress compressed chunk firstly please? Thanks

2

Hi @Yi_Sun , every data type will try a different compression algorithm. Unique values will also depend on the data type.

You can learn more how every algorithm works here: Time-series compression algorithms, explained