My Experience – SAP HANA: Columnar Storage vs Row Storage – Statistics on Compression

I want to know how much SAP HANA columnar storage table is efficient in terms of compression against row storage table and I have calculated these facts with my data. [I have data with 81,42,29,991 (more than 81 crores) rows and 10 columns in csv file and

memory size of the csv file is 81 GB.]

I have listed down compression statistics below for both row based and column based tables. [P.S: these facts and statistics with respect to only my data]

I have created both row table based and column based table using my data. Both the table is having 81 crore rows and 10 columns.

Row Storage Table:

Column Storage Table:

Both tables have same no of rows.

Compression:

We will check compression ratio in both row based table and column based table. [Note: Memory size of CSV file is 81 GB. Data is imported from this file ]

Row Storage Table:

Memory Size of the table is 74 GB (approx.)

Column Storage Table:

Memory Size of the table is 9.7 GB only (approx.)

Conclusion:

For my data, columnar storage compresses the data 7.6x times more than row storage. In column storage, the same set of data stored together and it gives better compression 🙂 .

Regards,

Muthu