SAP HANA - In-DB tools - cannot allocate enough memory
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Mute
- Printer Friendly Page
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Notify Moderator
Hi everyone,
I am currently running into an issue, for which I have not found a suitable solution in the Forum yet.
I have data in a SAP HANA table and want to build up a new workflow, which extracts data from this one (approx. 14mio rows and 100 columns) via a join and combines it with new calculated data (500k rows, 100 columns).
The workflow is completely set up with In-db Tools except for a small table that I am Streaming in via the "data Stream in" tool to filter by a join.
The problem is, that when I want to select the data from my historical cube and want to write it back to a temp table or combine it with new data, the full Memory of 64gb in HANA is blocked for this Operation before the workflow breaks completely….
The original database of 14mio rows is just 600mb on HANA and I cannot see any reason on why it is blown up to more than 40gb (24gb are already blocked before) temporarily during this Operation. At the Moment I cannot see any suitable way on how to deal wih this issue and if I can make any Settings on HANA or ALteryx side to avoid it....
Has anyone dealt with a similar Problem and can help me find a solution?
Any helpful advice appreciated! Thanks in advance!
Best regards,
Christian
Key Facts:
SAP HANA:
- 64gb tenant
- 24gb blocked before workflow runs
- historical cube approx. 600mb on HANA before workflow runs
User:
- Access to just one Schema
- read and write Access on HANA tenant
Solved! Go to Solution.
- Labels:
- In Database
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Notify Moderator
Is the join a 1-2-1 join? Or could you be duplicating records or even creating a Cartesian join between the two tables?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Notify Moderator
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Notify Moderator
Are both sides unique?
I'd be tempted (even though it may take more time) to run it using the standard tools locally so I have more visibility of the data and go from there.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Notify Moderator
Hi,
You have tried replacing from temporary table to to a physical table. I've had a lot of problems with popping out temporary space due to the way Alteryx works "With To"; I changed my processes to physical table and at the end of the process I delete them all.
[]
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Notify Moderator
The right side is just one row and joined via a left join. So for this join it just takes less than one second to proceed.
The issue really is the table to be written back to HANA at the end.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Notify Moderator
Yes @geraldo , unfortunately it is the same with writing back to a permanent table .... Have tried this as well.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Notify Moderator
Hi,
You can send a workflow print screen to view this join.
[]
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Notify Moderator
