Build better products with our product team
In many cases it is not needed to daily refresh the full semantic layer. It would be nice if TimeXtender supports the incremental refresh functions of Power BI Datasets to decrease load times and resource consumption.https://learn.microsoft.com/en-us/power-bi/connect-data/incremental-refresh-overview
There are tables like _L and _M to store the records when there are some errors in the load proccess.It would be nice to get a notification in the load proccess if these types of errors exists.Now the only way to know it it’s to see the content of the tables “manually”.
When moving data from DWH to a Semantic Model you get an interface where you choose the fields to take across and get the option to rename them. When you right-click drag to Data Areas you get the option to select fields but not rename them. As it is quite normal to want to rename many fields at the same time, it would be convenient to do this here instead of F2-ing or rightclicking the fields in the target Data Area one by one.
Instance variables cannot be used as value for Filter Row in an Ingest instance.This window lacks the option to use Instance Variables as a Filter Row, ie Date>=’VariableStartDate’This is possible in TimeXtender Classic with business units (Project Variables) and widely used at the customer. Currently Instance Variables seem only available on Prepare instances.We are using Dynamic Variables a lot for limiting the data load in Business Units in TimeXtender Classic.
When a table in MDW has multiple tables under Mapping (especially fact tables), but not all key fields are in all DSA tables, the default key should be set to value -1. In the current situation we have to add all missing key fields to the table in DSA and map them to MDW accordingly.MDW tables are in Simple Mode to reduce execution time. Disabling Simple Mode and setting a default value to each key field will work here, but that increases the time to populate the table significantly. Technically speaking, Default Value is a table constraint and not a transformation, which should make it possible to implement it on tables in Simple Mode as well.
In the current situation many steps for dimensions have to be done for each semantic layer separately, for instanceNew field in the dimension table in the MDW has to be added manually to all SSL models Descriptions on tables/fields/measures have to be filled manually for each model Idem for sort orders Idem for hidden fields Idem for hierarchies Etc.It would be a great time saver if some functionality exists where one canCreate a dimension based on a MDW-table Apply all properties, descriptions, sortings etc. Be able to reuse this dimension in multiple semantic layers Be able to apply changes to the dimensions to all semantic layers at once.https://www.kimballgroup.com/data-warehouse-business-intelligence-resources/kimball-techniques/dimensional-modeling-techniques/conformed-dimension/
For loading some data sources we rely on external triggers (eg. when files to import are ready on an FTP; when a record is added to a table etc). This is done by batch/PowerShell scripts via Windows Scheduler. When the files are ready, the corresponding TX packages will be started. This functionality is missing in ODX Server.It would be nice if ODX Server supports starting a task via either command line or PowerShell.https://legacysupport.timextender.com/hc/en-us/articles/210439143-How-to-Start-an-Execution-Package-from-the-Command-Line
Currently I’m using a set of fields that always I include them in all my MDW tables:It would be very useful can define a template of fields to include all that fields automatically in a selected table.
Creating the exact same dimensions in multiple semantic models can be a very tedious task. It is very common that models always share the same common dimensions. Having the ability to copy the structure between models or having some sort of templates would help with this. Having templates might also help with keeping changes for these dimensions in sync between models.
When implementing Custom Views you can map fields in the view output to an input parameter in the view syntax. This makes it possible for TimeXtender to track lineage through a view. There is currently a shortcoming in this approach: if you are combining two source fields from different tables into one field in the view, you cannot make a mapping for both. The only way to do this is to add a dummy field to the view that you do not actually need which wastes resources and reduces the clarity of the syntax.It would be nice if view fields can be mapped to at least one source parameter to resolve this.
In TimeXtender it is possible to create derived tables or subselections of tables which could benefit from automatically created covering indexes. Indexes are already created to drive Conditional Lookups but should also be possible for:Aggregate tables: especially if a small subset of fields are taken or if Data Selection rules are applied to the Aggregate Data movement: whenever only a portion of fields is taken across or Data Selection rules are appliedCovering indexes would make Execution faster and require less investigation effort once reloads start slowing down.Due to complete lineage being available for these TimeXtender automations, it should be possible to generate correct indexes. It is probably also beneficial to give the user the choice whether or not such an index should be added.
Hi community,The only way to know if a colleague has a work item on an item is to right click on it or to read the list of all work items. Since the deployment window is blocking, it is not possible to compare the tasks of a differential deployment with work items to make sure nothing is being overwritten. When an object gets added to deploy steps (either explicitly or implicitly), and it has a foreign work item on it, a warning should always be displayed.Kind regards,Andrew
Hi community, Currently the feature to create views using a SQL Snippet does not have the functionality to ‘Map Snippet View Fields', as do Custom Views. This means that all data lineage or impact analysis stops the moment a snippet view is involved. In projects with many snippet views, it means that data lineage is broken for the project.Kind regards,Andrew - E-mergo
Hi community, The query editor in ‘Manage Query Tables’ in ODX Server does not respect formatting. After saving and reopening all formatting is reverted. Before After:While this does not lead to an error in running the query, it makes it irritatingly difficult to edit and troubleshoot Query Tables, especially since the feature to run/execute a query table from the manager does not exist in ODX Server. Kind regards,ANdrew - E-mergo Support
It would be great to have an extra user role in the portal with the only permission to create new datasources and assign them to ODX instances. Currently if you want to give a user the permission to add datasources, they will also have the option to add, delete and change instances, which is, in some cases far from desirable.
The current TimeXtender Documentation documents list some metadata concerning what state of implementation is reflected. The filename contains a generation timestamp and the footer shows (usually) “latest deployed version” but sometimes “not the latest deployed version” which isn't very helpful.It would be good to have the actual version number listed in the documentation as “latest” is only guaranteed to be true at the moment of generation. This also makes it easier to track which documentation is for which state and environment.Being able to define what metadata fields end up in the filename and /or the footer would be even better.
Add option to create work items on Semantic Models, Semantic Tables and Semantic Model Measures.
We would like the following new filters in the ODX Execution log;In a drop down list, be able to select the data sources you are interested in seeing Be able to choose which status should be displayed. For example, show only failures. See the recipient in an outbound transfer. Before 6024 it will be environment, and in 6024 we want to see instance name.This would make it easier to find exactly what you are interested in.
When you set up execution packages in a chain (i.e. by triggering a package as a post execution step) and the execution of this chain causes a next scheduled execution to skip, currently the event log is not listing the name of the package that causes the skip.In other situations you will neatly be provided the name of the package, e.g. "The project 'TXProject' is currently executing the execution package 'AnotherPackage' in the process ‘timeXtender’”. However, with chained packages you will not be provided is name.When a chain of packages causes a skip of another scheduled execution, we would like to see the name of either the first or the last package in the chain listed in the event log.
The fields QuarterIndex, MonthIndex and WeekIndex must be relative to the current date without considering the quarter, month or week. It will be more usefull in PBI.Also consider to add DayIndex like:DATEDIFF(day,GETDATE(),[DateValue])
Aggregate Tables will currently show “Aggregate” in the DW_SourceCode field. In some cases users will want to pass on the DW_SourceCode present in the original table or perhaps specify one of their own.As system fields cannot be modified by transformations, changing this currently requires using a post-step or doing the aggregation in a Custom View or other feature.Features like Table Insert already allow you to choose how to fill DW_SourceCode, having a uniform approach across features would be more consistent.
Hi there! When you change a name of a DAX measure in TimeXtender, this always results in visuals crashing in the connected Power BI report when those visuals use the measures which names have been changed.It would be nice if Power BI could recognize the name change, without having to remove the measure with the old name from the visual and then having to add that same measure again to make the visual work again. Have a nice day! :)
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.