Time series Objects are increasing day by day

[pravin_tikar]

Hi Gurus,

In our project, we are facing one major issue regarding Time series objects are increasing day by day due to this Consistency chains are taking longer time and affecting the main chains to generate the required files on time.

We have monthly Unwanted MD delete chain but this is not able to delete the time series objects.

We have checked all master data in the system and it's not huge.

Please suggest your points of view.

regards,

Pravin

[Former Member · Accepted solution]

Pravin,

As your data grows, so will your TS objects.  It is normal for TS volume to increase in size, unless your business is very limited in the number of products, locations, orders, and time windows that must be supported.

Have you run the standard report to reorganize SNP Time series? 

SA38 > /SAPAPO/TS_LCM_REORG_SNP; use "repair".  Optionally, 'lock' if you think that updates will be taking place during execution.  This report will delete the TS for those MD objects that are no longer being used.

Best Regards,

DB49

[pravin_tikar · Accepted solution]

Hi DB49,

Thanks for reply.

We have executed the /SAPAPO/TS_LCM_REORG_SNP and /SAPAPO/TS_LCM_REORG_SNP_ALL.

Logs showing More than 10000000 Time series objects but not deleting...

Regards,

Pravin Tikar

[Former Member · Accepted solution]

Pravin,

If the reorg report is detecting many, but deleting none, that implies that none are superfluous.  I suppose it is also possible that the userid that is executing the report does not have adequate authorization to delete.  You can test this by physically deleting a Product Master using WUF, then running the reorg again.  The time series associated with that product should be deleted by the program.

Another thing to check are your /SAPAPO/TSINIT settings. When you run your weekly job executing this program, what are your standard settings?

Best Regards,

DB49

[pravin_tikar · Accepted solution]

Hi DB49,

I have checked the Times Series and Conc Step (Daily chain)

Setting are below for /SAPAPO/TS_LCM_REORG

Check for

Superfluous Master Data

Objects to be updated

Repair

Setting are below for /SAPAPO/TS_LCM_CONS_CHECK

PA,

Parallel Processing Profile:IN_10_1000

Checked

Repair

Check livecache anchor

Does this Parallel Processing Profile has any chance in not deleting Superfluous objects?

Regards,

Pravin Tikar

[Former Member · Accepted solution]

Pravin,

Did you test for authorization against //TS_LCM_REORG, as I suggested?

Did you look at transaction /SAPAPO/TSINIT (this is report /SAPAPO/TS_PAREA_INITIALIZE)?  what are the settings?  How often are you running this report?

Best Regards,

DB49

[pravin_tikar · Accepted solution]

Hi DB49,

Don't have any issue with authorization against //TS_LCM_REORG. we checked with BASIS Team.

We execute /SAPAPO/TS_PAREA_INITIALIZE Monthly chain.

Setting like below.

PA,

Parallel Processing Profile:IN_10_1000

Global Period.

As I already written, We have Monthly chain Unwanted_MD_Deletion.

Regards,

Pravin Tikar

[Former Member · Accepted solution]

Pravin,

I will assume you only have one Planning Version; 000; please confirm.

You should probably should also set "Update Master Data" for SNP PAs.

Start and end dates for Global period should be specific; generally one fills these with time variables.  They should only include dates that your planning solution will need.

Perhaps start date = 'today minus X days' and end date = 'today plus Y days'.  Use dates that are applicable for your planning solution.

After re-running this with restricted dates, again run the reorg report.

Best Regards,

DB49

[pravin_tikar · Accepted solution]

Hi DB49,

We have three Planning versions.000,001 & 002.

We have SNP Master data update sleected for SNP PAs.

Start & end date are maintain for Global period.

Please see below Variant setting.

regards,

Pravin Tikar

[Former Member · Accepted solution]

Pravin,

I don't know what is included in this "Unwanted MD Deletion" job, but I assume you are running //OM17 periodically and removing the superfluous product/locations and other superfluous entries from LiveCache.

I will also assume you are re-initializing all three Planning Versions during your monthly run.

Based on all of the statements you have made so far, and assuming that your Basis team actually knows what they are talking about, I would say that there is not much you can do to reduce the number of TS objects; without redesigning your solution.  As you accumulate more data you will get more TS objects, until the number of newly created objects equals the number that you reorg off the system each month.

Sorry & Best Regards,

DB49

[pravin_tikar · Accepted solution]

Hi DB49,

Yes, we are running Unwanted MD Deletion" job,  periodically and removing the superfluous product/locations and other superfluous entries from LiveCache.

Now we are doing one testing in QA.

We have removed the parallel processing profile and running the Time Series & Cons check.

It is taking so much time.... means it is identifying the superfluous entries.

Hope this will resolve our problem.

I will keep you updated..

Regards,

Pravin Tikar

[pravin_tikar · Accepted solution]

Hi DB49,

WOW...

We are able to delete the time series objects successfully in Q system based on parallel processing profile removing...

Now we have to test this in Production but after 1 months...

Keep you updated with result.

Regrads,

Pravin Tikar

[rahul_agarwal5]

Pravin bhai,

Is the cons chk taking longer time on a daily basis or only during the weekend/month end runs ? I hope this is only for SNP & not DP ?

Pl. confirm.

Regards,

Rahul

[pravin_tikar]

Hi Rahul Sir,

Cons check is taking time daily.only for SNP.

We have checked all possibilities to reduce the time series but not got any solution yet.

regards,

Pravin Tikar

[Former Member]

Hi Pravin,

Time series objects will grow because of following reasons:

1. Number of CVCs

2. Number of Key-figures

3. Initialization period of the planning Area

4. Number of versions

Please check the above four things and do the needful if you can reduce any of the above.

Regards,

Mukesh Pandey

[pravin_tikar]

Hi Mukesh,

Thanks for reply.

We have checked all above you suggested. but no improvement.

Regards,

Pravin