If you want to split a table on a single host SAP HANA system you would have to distribute the data over the numerous nodes for the table partitions. As soon as the time of partitioning a table, each of the partition acquires their own table-data structure internally, due to this each of the partition successfully performs like a table on its own. But for the fact you have to take care of which partition to be worked, that will depend up on the SQL Statement you are using.
Let us take an example i.e. the splitting by the age of the records. Consider that the maximum amount of data for which you are working and all the report that are related to the fresh days and that you hardly look at a data from a year ago. For such kind of situation you can make a partition i.e. one for every month. The fact here is that the data you don’t want to use would not need to be in memory all over the time. The same can be easily unloaded; by using this there would not be any kind of negative impact.
While performing the current partition only delta merge operations are to be used. At the time of execution of the query, many of the times the Hana optimizer, that you are using it for the recent data you would have to use WHERE condition for the splitting of the key. Please remember you can use too many partitions can be processed at the particular time, it will help you in the execution of the query much faster.
Hope this article will help you in splitting a table on SAP HANA system.