Table Partitioning

[Former Member]

Hi All,

I need some help in understanding table partitioning. I need to know what are the various considerations (like data volumes, type of operations - insert, update, select etc) that need to be kept in while deciding upon the partitioning type, number of partitioning etc. Thanks.

[former_member184768 · Accepted solution]

Hi Sanjay,

The information on this topic is widely available on this forum. Rather than explaining it once again, I'd point you to the documents / links.

Please refer to the document here to understand the concepts behind table partitioning.

Some threads;

http://scn.sap.com/thread/3334993

http://scn.sap.com/thread/3171886

This will help you understand partitioning better. If you have any more specific questions, please feel free to post it.

Regards,

Ravi

[Former Member · Accepted solution]

Thanks Ravi. I have been through these docs and they seem to be describing the concepts rather than answer my queries. Thanks.

[former_member184768 · Accepted solution]

Hi Sanjay,

If you understood the concepts, it will automatically answer your query.

As you might have understood, the partitioning depends upon the data volume. If your table is expected to contain data which can be logically partitioned on basis like time dimension, Geographical dimension etc, then depending upon the query pattern you can opt to partition the data based on these dimensions.

If you have multiple nodes, then HANA automatically partitions the data and stores on multiple nodes, but an explicit partition always helps from the query performance perspective.

No of partitions depend upon the grouping of the partition criteria.

In short, there is no straight forward answer to your question, but the factors like data volume, what data is stored (possible partition criteria), Query pattern etc decide how the partitioning should be done.

Regards,

Ravi

[Former Member]

Hi All, any views on this ? Thanks.