Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hi Sir/ Ma'am,
Good Day!
1. Is there a way to know the number of source tables on one task?
2. What is the best practice in creating a task, how many tables a task can handle?
Thanks, your response is highly appriciated
Hi,
Regarding your first question as you can see in the following screen shot at the bottom of the screen(marked in red) you can see the number of tables selected for replication. Does this answers your question?
Regarding your second question there is no known limitation on the number of table that can be replicated by one task. We have customer that are replicating few thousands of tables in one task. It all depends on the capacity of your machine. Here is a link to the Replicate documentation that lists the requirements for Replicate server machine:
https://help.qlik.com/en-US/replicate/November2021/Content/Replicate/Main/Installation/Recommended_H...
There are few parameters that may impact the resource consumption. If you wish you can contact your Qlik representative for getting further help onsite.
Hi,
Regarding your first question as you can see in the following screen shot at the bottom of the screen(marked in red) you can see the number of tables selected for replication. Does this answers your question?
Regarding your second question there is no known limitation on the number of table that can be replicated by one task. We have customer that are replicating few thousands of tables in one task. It all depends on the capacity of your machine. Here is a link to the Replicate documentation that lists the requirements for Replicate server machine:
https://help.qlik.com/en-US/replicate/November2021/Content/Replicate/Main/Installation/Recommended_H...
There are few parameters that may impact the resource consumption. If you wish you can contact your Qlik representative for getting further help onsite.
the other place to look for the number of tables is in the UI monitor Full-Load section. It shows those 'elevator' bars with a number beneath for completed/loading/waiting/error.
The same information, and more, can be obtained through the Enterprise Manager Rest API call 'GetTaskDetail' or the REPCTL call GETTASKSTATUS
The Api call GetTableList will (duh!) get list of the actual selected tables as does REPCTL GETTABLESSTATUS
As Orit indicates, there is no hard limit. A few hundred is certainly fine. But once you hit multiple thousands of tables the actions like initial task start and start-by-time-stamp can start to take long, from minutes to even an hours.
Also, you may need to find a balance in table activity. It may be desirable to have a single tasks for a handful of extra busy (CDC) tables and another task for thousands of low change volume tables.
Sample table count output structure below.
Hein.
full_load_counters:required structure {
full_load_completed_tables_count:required int32
full_load_loading_tables_count:required int32
full_load_queued_tables_count:required int32
full_load_canceled_tables_count:required int32
full_load_error_tables_count:required int32
full_load_total_record_transferred:required int64
full_load_estimated_records_count_for_all_tables:required int64