SAP NW71 queries

[Former Member]

Hi,

I have following queries regarding SAP NW7.1 application development.

1) Can we use all the data objects as Backend triggered while creating mobile application..what would be the pros and cons?

2) Can we configure DTR on NW7.1 server or it is not supported yet.

3) Can we have multiple user using the same device and share same database of mobile client?

4) Is it possible to filter data or send data on client based on Device instead of using any RMM like if multiple users are there using one device then distribution should to be on device based along with some other parametes like workcenter etc?

thanks in advance!!

Regards

Devendra

[christoph_kurtz2 · Accepted solution]

Hi Devendra!<br>

<br>

1. All data objects can be used as Backend triggered. The advantage is that in case of modifications of an object, these changes are sent directly to middleware system. In case of DOE triggered, the changes are sent to middleware system if one device is synchronizing with middleware.<br><br>

2. I think that the using of a DTR is not supported. You need a NWDI system. I have tried it with a DTR system and the result was that all my programs for 7.1 were deleted in NWDS.<br><br>

3. It is allowed to have more than one user on the same device. But I am not sure if they are using the same database.<br><br>

4. You can use a subscription generation data object to organize the data distribution. This special DO is only responsible for the distribution, so its content is not synchronized to any mobile device.<br><br>

Best regards,<br>

<br>

Christoph

[Former Member · Accepted solution]

Hi Christoph,

1. All data objects can be used as Backend triggered. The advantage is that in case of modifications of an object, these changes are sent directly to middleware system. In case of DOE triggered, the changes are sent to middleware system if one device is synchronizing with middleware.

Yes i know we can keep all dataobjects as backend triggered but is there any disadavange of using backend triggered object for all data objects. For example master data which is going to change once in a while we can keep it as backend triggered but there is no need for the same. Is there any performance impact or system memory consumtion issue ,Also when we make all DO as Backend push then in that case do you know if the Mobile Client sync will go to R/3 to do the delta or it will simply pick the data from Middleware without going to R/3?

2. I think that the using of a DTR is not supported. You need a NWDI system. I have tried it with a DTR system and the result was that all my programs for 7.1 were deleted in NWDS.

what do you mean by DTR system? Is it possible to setup DTR without setting up NWDI?

Do we need to configure a seperate server as NWDI..are there any patch/software available from SAP?

3. It is allowed to have more than one user on the same device. But I am not sure if they are using the same database.

4. You can use a subscription generation data object to organize the data distribution. This special DO is only responsible for the distribution, so its content is not synchronized to any mobile device.

yes thats correct but suppose i want to send all backend data specific to one user only to mobile client without customizing RMM when i have all the consolidated data for all users is available in middleware.

Thanks in advance!

Regards

Devendra

[christoph_kurtz2 · Accepted solution]

Hi Devendra!<br>

<br>

1. When you synchronize, the device is only connected with its outbound queue on DOE. It will not connect to backend. This is also the case if the data object is DOE triggered, but then, synchronization takes a longer time because DOE has to get the data from backend. If the DO is backend triggered, changes are automatically sent to DOE.<br><br>

2. DTR without NWDI is possible. But it is not recommend for MI 7.1 projects as I described in my last post. For more information regarding NWDI, please refer to help.sap.com, I am not an expert regarding this.<br><br>

4. I am not quite sure what do you mean by your post regarding the data distribution. If you define a subscription DO (responsible for data distribution), you don't need to customize RMM. RMM is only needed if you define a receiver generation DO (responsible for device generation).<br><br>

Best regards,<br>

<br>

Christoph

[Former Member · Accepted solution]

@Christoph:

1. When you synchronize, the device is only connected with its outbound queue on DOE. It will not connect to backend. This is also the case if the data object is DOE triggered, but then, synchronization takes a longer time because DOE has to get the data from backend. If the DO is backend triggered, changes are automatically sent to DOE.

Synchronization time is independent of DOE/backend triggered. Like you yourself said "When you synchronize, the device is only connected with its outbound queue on DOE". So when a device synchronizes, the DOE simply flushes the outbound queue of the device and thats it. So how the DO is populated into DOE from backend is irrelevant.

@Devendra:

yes thats correct but suppose i want to send all backend data specific to one user only to mobile client without customizing RMM when i have all the consolidated data for all users is available in middleware.

The USER field is part of the 'standard' device attributes shipped with the DOE. You can use it for 'Device Attribute Mapping' directly, without having to create a custom group.

3. It is allowed to have more than one user on the same device. But I am not sure if they are using the same database.

Yes. Its allowed. Correspondingly the shipped USER attribute in RMM is a multiple valued attribute. And the DB is common for all users. As per the DOE, data flows to 'Devices' and not 'Users'. Therefore the DB is at the 'Device' level and not the 'User' level. A 'USER' is just an attribute of the device which may or may not be used in the distribution logic. Nothing more.

Cheers!

[Former Member · Accepted solution]

Hi Arjun,

Thanks!!

I have one business scenario where we want to download the complete set of transactional data like Service order for a complete plant via getlist function module corresponding to data object for initial download and using BTE incase of any changes from R/3and then filter the data on middleware via distribution rule to seperate workcenters.

In all applications like MAM.MAU and other custom based applicaton i have seen that we use to filter service order data based on user id /userlevel in MI2.5?

what would be the problem if we use getlist to download the data for complete plant and use distribution rule like workcentre etc in NW7.1 to distribute data to devices? ( SAP is not supporting MAM3.1 in NW7.1? how it is being done in MAM3.1 do you have any idea?)

will there be any problem in future ( eg: like updating Service order or anything from client or queues running for long etc?)

Thanks & Regards

Devendra