HBase. How does Put in same row but different column/cell work?

Question

I am inserting data into HBase (0.96.1.1) by using multiple Put(s) to one and the same row-Id but different cells.

This will generate different (default) cell timestamps for each consequently put cell.

What impact is there on version-relevant stuff? e.g.

• are the Put(s) of different cells being counted towards the version limit in sense that when version limit is reached, the older Puts (even tho of different cells) will be removed/overwritten?
• are these cells affected by compactions again, in the sense that they will be removed?

My question concerns only the case when Put(ing) different Cells into same row - other cases are clear.

Answer 1

A {row, column, version} tuple exactly specifies a cell in HBase. Since version is per column, each column will maintain its version count and only versions of those columns are eligible for removal which reach version limit.