SAP for Utilities Discussions
Connect with fellow SAP users to share best practices, troubleshoot challenges, and collaborate on building a sustainable energy future. Join the discussion.
cancel
Showing results for 
Search instead for 
Did you mean: 

Energy Data Management

former_member199650
Participant
0 Kudos

Hi,


We are facing performance issue once using the BAPI (BAPI_ISUPROFILE_IMPORT).

Please let us know, how can we find out std. enterprise services? Please suggest…..

Is there any benefit in using these standalone enterprise web services. Can we reduce the performance/tuning

related issue with the help of these two enterprise services? 

1.  UtilitiesTimeSeriesERPItemBulkCreateRequest_In

2.  UtilitiesTimeSeriesERPItemBulkChangeRequest_In

Thanks and Regards

MP Vashishth

3 REPLIES 3

Former Member
0 Kudos

Hello Mahavir,

actually I haven't understood the issue and what you're expecting from calling the mentioned web services.

You should take into account that for a web service, the data needs to be converted into XML payload, which could become quite bulky compared to the RFC call. So using a web service could make things worse.

The performance of the Profile Values import depends on the total number of records, which could be reduced by using smaller date ranges (e.g. 6 months instead of 2 years) and using aggregated values (e.g. 1 hour instead of 15 minutes). Whether and how this is applicable depends on your scenario.

Best regards,

Christian.

0 Kudos

Perhaps the following document is helpful, it contains several recommendations:

https://service.sap.com/~sapidb/012002523100018701262014E/EDMImportExport.pdf

Message was edited by: William Eastman

Former Member
0 Kudos

Hi Mahahir,

1) To address the issue with BAPI_ISUPROFILE_IMPORT, here are a few questions you will need to ask yourself:

- Are you passing the data in using the BAPI is one large call or are you breaking up the data into smaller packets? In previous implementations I have worked on, this is the typical way to ease performance issues by using parallelisation.

- How much data are you trying to load (i.e. 1 day of 15 minute data for 100 profiles = 9600 individual values)?

- What other jobs/programs are running at the same time? Are there semaphores applied to any processes when checking SM50?

- Have you checked the performance/configuration of the system with your BASIS team? Especially using ST02, DBACOCKPIT etc...

2) To address the issue with using AMI Enterprise Services:

- AMI web services are a change in technology, not a performance booster. Typically these are used when communication with an external system  has the capability of communicating using web service calls directly from system to system using XML.

- AMI can offer customer enhancements up front when the data is received (i.e. extra validations) before EDM even touches the data. Its a nice touch when performing some validations up front.

- If your system does not have the appropriate resources (i.e. cores, RAM etc), then as Christian has stated, this is probably not a good idea to try this out.

Regards

Zane