Using Power BI against Infor Data Lake in 2024?

I work at a business that has a history of using Power BI against their ERP system for dashboards and reporting (created in Desktop and consumed online), also using a custom data warehouse solution, and April 2023 they went live with Infor CSI. Power BI is still used almost elusively, however Data Views are sent as CSV files via SFTP, and then Power Automate flows to perform ETL.

The problem since the change to Infor CSI is the lack of near real-time data, with the Data View export occurring as a Background Task once a day. However, I recently joined the business and have got Data Lake up and running, but the challenge is now how to consume this data in Power BI.

As far as I'm aware JDBC is not possible as Power BI does not support it, so I'm left with two potential paths, 1: Compass API, 2: get the Infor Data Lake data into Microsoft One Lake.

Honestly I'm drawn to Microsoft One Lake, as if the transfer of data works well we carry on easily using Power BI and our reporting platform is not locked into a single ERP vendor. However, this may be infinitely more difficult to achieve than using the Compass API and using this to pass SQL queries from Power BI.

I suppose I'm asking for miracles in that there's someone here who has either been through the same situation or can provides insightful comments.

Thank you,

Clive.

Edit: Based upon replies from other Infor users it seems like there are two potential choices... 1: Use the Infor ETL Tool / 2: Under Infor OS > Data Fabric we have the Pipelines feature, which over time will have other "destinations" added to it and hopefully these will include things such as MS SQL Server and MS One Lake.

Find more posts tagged with

General Discussions

Accepted answers

Bojan Rafajloski

Hi Clive,

I can confirm that we are targeting 2025 (tentatively - as with any of our roadmap) to support Azure SQL and Amazon RDS SQL Server as destinations in Stream Pipelines. Currently, Stream Pipelines supports Azure Postgres and AWS Postgres destinations, but as said, we plan to expand the connectivity options in 2025.

You are not seeing Stream Pipelines in your tenant because it is an additional service and is not included in the standard OS Platform licensing.

Stream Pipelines are designed for real-time data delivery, however that can be only be achieved when the source is also streaming the data directly to Data Fabric. In such scenarios it utilizes the Data Fabric streaming architecture to provide the quickest path for the data events end-to-end, eliminating any stops in data movement.

In the other scenario, when the source application is publishing data through ION or to Data Lake, pipelines will process the data after it is stored in Data Lake. So depending on the source replication schedule, that's how often you can expect data to be fresh.

The ETL Client for Data Lake is a on-prem installation that you can use to schedule data extraction from the Data Lake to perform incremental load with upserts to a SQL database. ETL Client is also an addon license that is not included in the OS packaged licensing.

You can find more information in our user guides, Infor U Education and our YouTube channel:

ETL Admin Guide
Stream Pipelines
Data Fabric Extraction
YouTube playlist for Infor Data Fabric
In Infor U Education search for keyword 'Data Lake Part' and the result should be 3 part sessions on Data Lake topics, including extractions and the ETL Client

Please contact your account representative and they will be able to help discuss further your use cases and possibilities.

All comments

Andrew Sowden

We're starting the journey using MS Fabric too .. I will follow this thread with interest

kvargas

Hi Clive

We are in FSM and HCM implementation right now and came across the same issue. We chose to purchase Dashboard Gear which is essentially a ETL tool to an on-prem SQL server data store. The solution utilizes Infor Replication sets to move the data. Dashboard Gear 100% manages the ETL and any requests to create custom calculations etc. It is a delivered nightly run, but you can request selected business classes to run more frequently. The solution also provides cubes and Power BI data sets in addition to the SQL server tables. Dashboard Gear set this up in less than a month and some of that time was getting the SQL server database set up and permissions granted.

tdouglass

We are Multi Tenant Cloud M3 and are unaware of any way to connect without using the Infor ETL tool to extract to SQL Server. There will be a streaming to SQL server available in 2025. We could also possibly do the API calls to feed into SQL Server.

Daniel Turner

Another option is to use a JDBC to ODBC bridge driver. We've used one from OpenLink software. It works, but speed is affected. As a very non-scientific test, I grabbed all of our data from ICSW using the JDBC driver directly in DBeaver and it took about 20 seconds to fully load 27,000 rows x 234 columns. Using the ODBC driver as a bridge to JDBC, it took about 39 seconds. I've not tested it with Power BI, but it does work as a data source in excel, so my assumption is it would also work with Power BI. I think they have a free trial on it. The setup is a little clunky on it, but once it's up and running, it's pretty hands off. uda.openlinksw.com/.../

christophedaeninck

We moved from iSeries to the cloud, so we had a good knowledge of the tables and used datastructure. To skip the investment of redeveloping interfaces towards several producton application and the QlikView dashboards we bought the Infor ETL. The ETL tool synchronizes several (140) tables to a MS SQL server. The tables are grouped into 3 possible levels of synchronization: every 5 minutes, every hour, and once a day. The only thing we had to change during migration was updating the ODBC connection of the apps, BI and excel files. Actually we are also using this ETL SQL database to synchronize with an Azure BD to start developping PowerBI dashboards. With the delay of 10 minutes between M3 and DataLake, adding the 5 minutes from the ETL tool we can get real time data with only a 15 minutes delay.

unknown

I've not yet looked at the Infor ETL suite / tool, was this not suitable for you so you used a third party one?

unknown

"There will be a streaming to SQL server available in 2025"... well, that would be perfect, especially if we can use it to stream to Azure SQL etc.

Do you mind me asking how you know this, is it on a road map that's publicly available?

Edit: I've just noticed that under Infor OS > Data Fabric I'm missing the "Pipelines" option, so I've logged a support case for this. I'm assuming, that in the coming months new destination for Pipelines will be added with MS SQL Server being possibly one of them.

unknown

Thank you.

A next step for me to start looking at Infor's ETL tool.

Is the ETL tool an actual Windows / Linux application or is an on-line one?

christophedaeninck

It is a JAVA package, so it should run on a Linux environment.. We have it installed on a dedicated Windows server and it uses an MS SQL server as repository to store transformation.

Bojan Rafajloski

Hi Clive,

You are not seeing Stream Pipelines in your tenant because it is an additional service and is not included in the standard OS Platform licensing.

You can find more information in our user guides, Infor U Education and our YouTube channel:

ETL Admin Guide
Stream Pipelines
Data Fabric Extraction
YouTube playlist for Infor Data Fabric
In Infor U Education search for keyword 'Data Lake Part' and the result should be 3 part sessions on Data Lake topics, including extractions and the ETL Client

Please contact your account representative and they will be able to help discuss further your use cases and possibilities.

unknown

Ding ding ding, we have a winner

I've seen a LinkedIn post you made, along with Infor videos on YouTube, that steered me to Stream Pipelines. I've got the Data Lake side of things cracked from Infor CSI, so it was the step to get it easily into Power BI that was missing.

I now have a path though to follow... get costs from my account manager for Pipelines > wait for Pipelines to get its new Destinations.

Is there a KB / YouTube channel / other source that I should follow to know when new Pipeline destinations are announced?

Thank you.

Roy Es van

We developed a tool to replicate the data from the datalake. The data is send trough a datalake flow to a enterprise connector location or SFTP and the files are imported by the tool. The tool will also read the documents (tables) connected to the datalake flow and create the tables and columns automatically. At the moment windows and sql server are supported. If you want more information, don't hesitate to send me a message.

Copyright © 2025 Infor. All rights reserved.