Parallelized transfer

[Former Member]

Hi

Can any one explain what is parallelized transfer in the screen CFM2 transaction(activation or deactivation integration model). What is the use of this and how we can use for better performance.

What must be the values for

1.Absolute Max no of process

2. Relativemax no of process

Please share if any one has used this option.

Thanks and regards

VIjaya

[Former Member]

Hi Vijaya,

When we carry out integration model run, we used to carry out transfer of objects sequentially.

This option is useful to carry out paralleized transfer of more than one or more objects at a time.

If you set this indicator, the ERP system selects objects that are to be transferred to SAP APO in parallel in different work processes. The parallelization takes place with reference to the blocks of a filter object type. Each work process processes a filter object block.

You specify the block size (that is, the maximum number of filer objects to be processed per block) using transaction CFC3 as before.

Parallelized selection can lead to a significant increase in performance.

If this indicator is not set, the ERP system selects the objects in serial.

The performance of the initial data transfer is not only determined by parallelization, but also by the size of the filter object blocks. If you specify a large block size in transaction CFC3, only a few blocks may be formed within the filter object type. This means that only a few work processes can be parallelized.

However, for smaller block sizes it is possible to have a higher degree of parallelization although only a small number of objects are transferred to SAP APO per block

If you would like to optimize the performance of the initial data transfer, you also have to take into account the size of the filter blocks in addition to parallelization and adjust your system depending on the data being transferred.

In general, in most of the projects, they prefer sequential transfer rather than parallel transfer in which there will be some uniformity in transfer process and any errors caused during transfer can be identified and actions can be taken online for those objects. If you transfer all in one shot, error analysis will be very challenging one.

Regards

R. Senthil Mareeswaran.

[Former Member]

Hi Senthil,

Thanks for your response.

We have deactivated the GATP integration models for last 3 weeks due to some user mistake. The ATP is with in R3 only. Now we have resolved the issue and started activating all the models. But the jobs are canceled.There are huge no of orders accumulated in R3. So we thought to use this process temporarily.

We have done required changes in CFC3. Now it worked.

My question is answered now.

Regards

Vijaya

[Former Member]

Hi Vijaya,

When we carry out integration model run, we used to carry out transfer of objects sequentially.

This option is useful to carry out paralleized transfer of more than one or more objects at a time.

If you set this indicator, the ERP system selects objects that are to be transferred to SAP APO in parallel in different work processes. The parallelization takes place with reference to the blocks of a filter object type. Each work process processes a filter object block.

You specify the block size (that is, the maximum number of filer objects to be processed per block) using transaction CFC3 as before.

Parallelized selection can lead to a significant increase in performance.

If this indicator is not set, the ERP system selects the objects in serial.

The performance of the initial data transfer is not only determined by parallelization, but also by the size of the filter object blocks. If you specify a large block size in transaction CFC3, only a few blocks may be formed within the filter object type. This means that only a few work processes can be parallelized.

However, for smaller block sizes it is possible to have a higher degree of parallelization although only a small number of objects are transferred to SAP APO per block

If you would like to optimize the performance of the initial data transfer, you also have to take into account the size of the filter blocks in addition to parallelization and adjust your system depending on the data being transferred.

In general, in most of the projects, they prefer sequential transfer rather than parallel transfer in which there will be some uniformity in transfer process and any errors caused during transfer can be identified and actions can be taken online for those objects. If you transfer all in one shot, error analysis will be very challenging one.

Regards

R. Senthil Mareeswaran.

[Former Member]

Hi Vijaya,

When we carry out integration model run, we used to carry out transfer of objects sequentially.

This option is useful to carry out paralleized transfer of more than one or more objects at a time.

If you set this indicator, the ERP system selects objects that are to be transferred to SAP APO in parallel in different work processes. The parallelization takes place with reference to the blocks of a filter object type. Each work process processes a filter object block.

You specify the block size (that is, the maximum number of filer objects to be processed per block) using transaction CFC3 as before.

Parallelized selection can lead to a significant increase in performance.

If this indicator is not set, the ERP system selects the objects in serial.

The performance of the initial data transfer is not only determined by parallelization, but also by the size of the filter object blocks. If you specify a large block size in transaction CFC3, only a few blocks may be formed within the filter object type. This means that only a few work processes can be parallelized.

However, for smaller block sizes it is possible to have a higher degree of parallelization although only a small number of objects are transferred to SAP APO per block

If you would like to optimize the performance of the initial data transfer, you also have to take into account the size of the filter blocks in addition to parallelization and adjust your system depending on the data being transferred.

In general, in most of the projects, they prefer sequential transfer rather than parallel transfer in which there will be some uniformity in transfer process and any errors caused during transfer can be identified and actions can be taken online for those objects. If you transfer all in one shot, error analysis will be very challenging one.

Regards

R. Senthil Mareeswaran.

[Former Member]

Hi Vijaya,

When we carry out integration model run, we used to carry out transfer of objects sequentially.

This option is useful to carry out paralleized transfer of more than one or more objects at a time.

If you set this indicator, the ERP system selects objects that are to be transferred to SAP APO in parallel in different work processes. The parallelization takes place with reference to the blocks of a filter object type. Each work process processes a filter object block.

You specify the block size (that is, the maximum number of filer objects to be processed per block) using transaction CFC3 as before.

Parallelized selection can lead to a significant increase in performance.

If this indicator is not set, the ERP system selects the objects in serial.

The performance of the initial data transfer is not only determined by parallelization, but also by the size of the filter object blocks. If you specify a large block size in transaction CFC3, only a few blocks may be formed within the filter object type. This means that only a few work processes can be parallelized.

However, for smaller block sizes it is possible to have a higher degree of parallelization although only a small number of objects are transferred to SAP APO per block

If you would like to optimize the performance of the initial data transfer, you also have to take into account the size of the filter blocks in addition to parallelization and adjust your system depending on the data being transferred.

In general, in most of the projects, they prefer sequential transfer rather than parallel transfer in which there will be some uniformity in transfer process and any errors caused during transfer can be identified and actions can be taken online for those objects. If you transfer all in one shot, error analysis will be very challenging one.

Regards

R. Senthil Mareeswaran.