on 01-20-2020 12:11 PM
Hi,
We want to copy 16 billion records from one table to another table in HANA DB. At the moment it takes 6 hours to copy 2 billion records. At this rate we might need couple of days to complete this activity. Is there a faster way to achieve the same?
Our table are range partitioned and we are using following code:
INSERT INTO "SAPABAP1"."TABLE1" ( SELECT * FROM "SAPABAP1"."TABLE1_BCKUP" );
we are on HANA 2.0 and DB version 30 and SP03.
Thanks
Just like with your other mass-data-processing question, the single most important thing with performance tuning is to figure out what is happening during the (obviously too) long runtime of the statement/program.
Based on your description we can assume that the time reading the data is probably happening as quickly as possible (no join/aggregate computation).
A common issue with huge data loads is the default handling of delta merges with column store tables.
The topic is not overly complicated but probably too much to explain in detail here.
I've written a blog post on that before, so why don't you go and have a look:
https://lbreddemann.org/merging-right-some-tests-with-data-loading-in-hana-from-down-under/
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
User | Count |
---|---|
80 | |
9 | |
9 | |
7 | |
7 | |
7 | |
6 | |
5 | |
5 | |
4 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.