Quantcast
Channel: SCN : Discussion List - SAP HANA and In-Memory Computing
Viewing all articles
Browse latest Browse all 5653

Select is slow on Scale-Out HANA db for BW

$
0
0

Hi Gurus,

 

I need some help...

 

We have a 3-tier BW on HANA landscape. The Dev and Test have a scale up. The Production is a scale out with 4 nodes. 1 master with 3 slaves.

We went with scale out as we were told you cannot go scale up if you need more than 1TB.

 

We have a transformation which includes Select statement that joins on a DSO active table (partitioned across 3 slave nodes) and a temporary db table (on master node). This select is running significantly slower in production as compared with Test system. Test takes between 1 - 3 hours and production takes 30 hours. The same code and same amount of data. The only difference is the hardware architecture.

 

The Select statement is run few hundred thousand times in a loop hence the performance of the select needs to be optimum. This is a user requirement and cannot be changed. Basically it is traversing through a hierarchy (parent child records).

 

            SELECT dso~H_NODEID as NODEID dso~H_PARENTID as PARENTID
            dso~/BIC/HSID as HSID dso~/BIC/ABC as ABC

            dso~/BIC/ABCTY as ABCTY dso~/BIC/ABC as ABC_PARNT
            INTO TABLE lt_children
            FROM /BIC/AXP_O0500 as dso
          INNER JOIN ZTEMP_1 as parent
                ON  parent~NODEID = dso~h_PARENTID
          WHERE h_hienm = p_hienm
                 AND  dso~/BIC/ABCTY NE 'C'

 

Your Help is appreciated.

 

Best regards,

AG


Viewing all articles
Browse latest Browse all 5653

Trending Articles



<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>