SAP recommends using compression on column store databases to conserve RAM and disk space resources (https://blogs.sap.com/2016/08/12/how-and-why-to-do-compression-of-column-store-tables-in-sap-hana/). Since date+time values are typically unique and compression in SAP HANA is column-based, compression of timeseries data like the data published by the Integrator is more effective if date and time are in separate columns as opposed to a single time column containing both date and time.

Native ArcGIS Feature Services allow for features to be updated by performing a POST to the REST endpoint to "Apply Updates".  It would be good to be able to record new values to the PI Data Archive by "Applying Edits" to the features / Elements made available for read access by the Integrator Hosted Time Enabled Feature Services.

I wish that the start time of a continuous view update cycle could precede the end time of the last cycle.

For example every day query the last 30 days of data and replace rows that have already been written to the target table with the most recent data. This would be a straightforward way to ensure that data in PI and the integrator View are the same for the last 30 days. Manual corrections up to 30 days old could then be captured automatically.

Update the naming convention used to create ArcGIS feature services and layers to use names that more accurately reflect the services and layers stored on the Integrator side. Customer Issue: The current convention is to use numbers instead of the layer names which causes confusion when trying to locate published items in ArcGIS Online or Portal for ArcGIS.

The PI Integrator for Esri ArcGIS does not support SAML authentication through Portal. Support for SAML authentication would allow PI Integrator users to configure their own Identity Provider.

The backfill option in Azure IOT integrator should handle the raw data, instead of pushing the 1 minute interpolated values.

There are automatic updates for PI data in Asset Views and Streaming Views, but not currently in Event Views which is what's used for Event Frames. This is important for propagating changes in EFs

Use case example: GADS classification. Data comes in and EFs are created. The integrator picks this up and sends the info along. Some time later (maybe a few days / maybe month end) data comes in re-classifying / updating the original data (ex. GADS code or attributes). In order to accommodate this manual intervention using the "Update Data" button is required.

The Integrator currently supports only a single SQL server connection string to access all of its backend databases (PIIntegratorDB, PIIntegratorLogs, PIIntegratorStats). Our organization's SQL Server cluster provides high availability at the SQL database level, however, with a separate DNS listener for each database. When either of the other PI Integrator backend databases fails over to another SQL server, we see errors like the following:
“SqlUtilities:GetBackendSqlConnection: Error: Unable to obtain open SqlConnection to database, ‘PIIntegratorLogs’,at SQL server path, '(Server)'".

I would to force the number of decimals the Integrator writes for a given analog value

Have an extra column in the view that indicates when the data row is actually written to the target.
The current timestamp column is the timestamp of the value itself, not when the data row is written.
This will be especially useful for troubleshooting, or when PI Integrator is updating existing value in a target table (will give indication of which data row was recently updated).

Improve performance of queries resulting in a narrow table. This can happen for queries with many Elements, each with a few Attributes, where bulk queries to the data source are not possible. A narrow table can also result when using * for the Attribute criteria, as each Attribute is then considered a separate match.

As a PI Integrator Administrator I need to grant permissions to a group of users to configure targets since they are the administrators of the target systems (SQL Server, ADLS, etc.), but they don't need access to other administrator tasks like security management.

When there are multiple attributes named the same for different elements, the full path is automatically done to avoid duplicated column names. Element without name duplication issue should also have the option to display the full path column name.

As a PI Administrator for a power company I am asked twice a month to verify that none of the products I support have had a security patch released.  Any such products in a NERC zone need to be patched within the next 12 days.

It used to be fairly easy to look at updates to all PI Client tools to find released security patches.  So far I haven't been able to find out if any have been patched.  This would also include interfaces which are prone to being run in a NERC environment.

To execute a view with the same parameters, it is actually mandatory to edit the view to be able to execute it again.
It would be nice to be able to force its execution using a right clic on it for example

When applying a filter when building a shape, the Integrator displays a limited amount of matching elements which is fine.
It would be useful do display the number of element matches as an additional info

The PI Integrator should validate any link entered in the endpoint section (PI Vision, Portal, GeoEvent). It should warn the user of any certificate issues or port that are not opened.