TM1 CubeUnload

Post Reply
cipher
Posts: 13
Joined: Thu Jan 08, 2015 6:15 am
OLAP Product: TM1
Version: 9.5.2 10.2.2 64 Bit
Excel Version: 2010

TM1 CubeUnload

Post by cipher »

hi All,

We use 10.2.2 with fix pack 6 and excel is 32 bit office 365. Server is hosted on 2012 R2.

I am trying to understand usage of CubeUnload function. I am struggling to understand in which scenario I should be using CubeUnload() function.

Whenever I perform CubeUnload on any cube, TM1Instance or TM1sd.exe doesnt release that memory to OS. The RAM consumption by TM1 remains as it is. I am assuming that with CubeUnload() function TM1 takes off space occupied by cube and that space can be re-used by TM1 to perform rule calculation or open views. I am not sure if thats the case. I need your views on it.

Currently our model is occupying 90% of total ram. What should I do to reduce memory footprint of TM1 Instance except deleting unnecessary cubes?

Also, I need to know the any scenarios where CubeUnload can be useful.

Regards!
User avatar
orlando
Community Contributor
Posts: 167
Joined: Fri Aug 04, 2017 8:27 am
OLAP Product: TM1
Version: PAL 2.0.8
Excel Version: Office 365

Re: TM1 CubeUnload

Post by orlando »

Hi,

i use CubeUnload if it is necessary to restore just one cube and i don't want to restart the hole tm1 db.

If you want to save RAM, I don't think this will help. Therefor you have to check your rules and feeders - maybe they consume to much RAM.
Check your sturctures - maybe you can reduce them etc..
reorganize the order of the dimensions in your cube (i always to this via the assistant - and it helps)

regards,
orlando
David Usherwood
Site Admin
Posts: 1453
Joined: Wed May 28, 2008 9:09 am

Re: TM1 CubeUnload

Post by David Usherwood »

TM1 never hands back memory to the OS.
If you can identify cubes which are not required and also not linked via rules or feeders to cubes you do need, you can set these to Load on Demand and reduce the RAM footprint.
Otherwise I agree with orlando - review your feeders using the Performance Monitor cubes, and look at resequencing your large cubes. You may need conditional feeders.
Paul Segal
Community Contributor
Posts: 306
Joined: Mon May 12, 2008 8:11 am
OLAP Product: TM1
Version: TM1 11 and up
Excel Version: Too many to count

Re: TM1 CubeUnload

Post by Paul Segal »

Or, buy some more memory ;)
Paul
User avatar
Elessar
Community Contributor
Posts: 332
Joined: Mon Nov 21, 2011 12:33 pm
OLAP Product: PA 2
Version: 2.0.9
Excel Version: 2016
Contact:

Re: TM1 CubeUnload

Post by Elessar »

If you turn on Performance Monitor and open }Statsbycube cube, you'll see measures:
Memory Used
Memory In Garbage

When you use CubeUnload, the memory used by cube will moved from "Used" to "Garbage". As David said, it will never be released to OS, but it can be used inside TM1 for creating stargates and so on.
Best regards, Alexander Dvoynev

TM1 and Data Science blog: 6th article - PAfE + VBA: Commit each cell without pressing “Commit” button.
Wim Gielis
MVP
Posts: 3105
Joined: Mon Dec 29, 2008 6:26 pm
OLAP Product: TM1, Jedox
Version: PAL 2.0.9.18
Excel Version: Microsoft 365
Location: Brussels, Belgium
Contact:

Re: TM1 CubeUnload

Post by Wim Gielis »

For example, you would use CubeUnload (or the equivalent in the user interface) to remove all feeders from a cube. Useful during debugging. Technical requirement: all views must be closed in Architect / Perspectives.

This is not a tool for limiting RAM consumption in a TM1 model. For that, you have to spend more time and review the contents of the rules files (*.rux).
Best regards,

Wim Gielis

IBM Champion 2024
Excel Most Valuable Professional, 2011-2014
https://www.wimgielis.com ==> 121 TM1 articles and a lot of custom code
Newest blog article: Deleting elements quickly
Post Reply