The Infor Communities is globally accessed system by
Infor and its customers, please do not post or upload
any content that could be deemed as sensitive, confidential or
subject to any data regulation requirements.
If you are unsure whether or not the content contains sensitive,
confidential or has any data regulation requirements,
please seek the advice of your internal security, legal, and compliance experts.
[How To] Load data from a database to the Data Lake by Mike Kalinowski
We're seeing an exciting increase in the number of users and customers starting to use Data Lake and pre-built content provided by Infor applications and pre-built content for applications like Birst. We'll start introducing a series of HowTo guides through the forums, blogs, and our Infor OS YouTube channel. If you're interesting in learning more or have a request for new content, reach out to me through the forums and we'll see what we can do to stand up content for the community!
For our first guide, let's learn how to use Infor ION to connect to and extra data from a database and use the Data Lake's Compass Editor to query that data.
Before we jump in - you can follow along by clicking on any of the GIF images in the post to see a demonstration!
Let's begin with Infor ION
Infor ION comes with a suite of technology connectors so that you can connect to databases, file servers, APIs, JMS queues, and more. We'll need to set up a database connection point so that we can start modeling our AnySQL data extraction. Before you get any further, make sure you've already set up the ION Enterprise Connector so that Infor ION in the Cloud can access your on-prem database.
Navigate to ION Desk and select the Connect menu option to expand and drill down to Connection Points. Click the Add tile and select the Database option. You'll need to fill in your database's connection details including:
Our next step is to drill down into the AnySQL modeler. Before you get too far, make sure that you've selected AnySQL Modeler in the Connection Point's details pain under the Type drop-down. Select the Documents tab and add a new blank model to the datagrid and hit the Edit button to launch into AnySQL.
AnySQL is a clever tool that uses your connection point details to provide a graphical modeler where you can preview, drag & drop tables, and generate metadata that represents your database table you'll query in the Data Lake.
Expand the data discovery panel by selecting the + icon in the modeler and either preview or drag and drop your tables into the modeler. If you need to join tables together, simply drag a new table over top any existing table in your model. Make sure to select the Columns icon for each table in your model and pick which columns you'll want to capture.
Once you're finished, you can preview the AnySQL model's output or SQL syntax - or simply save your model and click the GENERATE METADATA button where a copy of your object metadata will be registered into the Data Catalog.
Setting up our Data Lake Flow
Using ION's familiar modeling tools for data movement, we'll navigate to the Connect menu and drill down to Data Flows where we can click on the Add tile and select the Data Lake Flows option.
Data Lake Flows are an explicit modeling option for moving any integration files you might be using in Document Flows to also send a copy over to the Data Lake. Data Lake Flows can be used to both ingest data into the Data Lake as well as extract data from the Data Lake and send it to another connection point registered in ION. For the latter scenario, that's typically used for batch-based integrations.
Once we're in our flow modeler, drag the Database activity from the Activity Toolbar and drop it into an empty space in the model. Select the drop-down in the details panel and select our database connection point we used earlier. Next, drag the Ingest step and add it to the flow after our connection point. Click on the document icon between our two activities to add our AnySQL model from the Data Catalog we've created earlier.
Now - click Activate on your flow and you're pumping data into the Data Lake!
Querying our data
Once you've activated your flow, you can monitor data extractions through the OneView utility within ION Desk. In the example below, we're using basic filters to find data sent in the last hour but you can more complex and advanced filters to find the data you're looking for. Once you see that your object has been sent to the Data Lake, let's go ahead and visit the Data Lake's Compass Editor to query it.
Navigate to the Data Lake menu and drill down to Compass. Here, you'll see a list of data objects sourced from the Data Catalog. We've done a quick filter for objects that contain the word "Community." In the editor, you can your own SQL query or quickly generate one by hovering over a table and selecting the ... icon and drilling into the SQL sub-menu under Generate and selecting one of the options available.
And that's it! You're able to quickly and easily model a data extraction and use ION to ingest that into Data Lake so other users, applications, and systems now have data to query against.
RE: [How To] Load data from a database to the Data Lake by 1314892: Hi @mike, Thank you for the information. I am trying to create a database connection point but not getting anything under Location. Do I need to run the agent to get the list of Locations?RE: [How To] Load data from a database to the Data Lake by Mike Kalinowski:
You'll need to download and configure the ION Enterprise Connector somewhere within your network. It's a small, light-weight agent that allows ION to communicate and exchange data from on-prem systems to the cloud and vice versa.
You can find more information in Chapter 5 of the Infor ION Desk User Guide - roughly page 54+ :)
RE: [How To] Load data from a database to the Data Lake by 1314892: Hi Mike, Can we embed Birst into On-Premise OS ? If Yes, How can we load the LN data to Birst? I believe there is no Data Lake in On-premise ION.RE: [How To] Load data from a database to the Data Lake by 1314892: Can you please help me on this.RE: [How To] Load data from a database to the Data Lake by Kevin Heiman: Correct, Data Lake is not available for on-premise. The option would be to use a Hybrid of Cloud and On-premise, and leverage the Cloud services as needed. Enterprise Connector is also only for Cloud environments to private external networks as Mike noted above.RE: [How To] Load data from a database to the Data Lake by 1314892: Hi Kevin, Thank you for the response. As per my understanding, we cannot use datalake and enterprise connector in on-premise. Can you please correct me if I am wrong and explain in detail. Thanks, Rajini.RE: [How To] Load data from a database to the Data Lake by Mike Kalinowski: Hey Rajini, Data Lake is only available to Infor OS MT Cloud environments. Enterprise Connector would only be utilized insofar as connecting to and pulling (or pushing) data from LN on-prem to the Cloud. However, I believe that's an unsupported configuration in the LN world. If you're running on-prem, then you can connect Birst directly to LN itself, I believe. I would recommend checking the documentation portal to see if there's more detail here or else, reach out to your account manager for more details. ThanksRE: [How To] Load data from a database to the Data Lake by 1314892: Hi Mike, Thank you for the response. I read that we need to download and extract the BIRST SAML Signing Certificate for embedding into on-premise birst. Can you please help me from where to download the Birst SAML Signing certificate? Thanks in advance. RajiniRE: [How To] Load data from a database to the Data Lake by Mike Kalinowski: Hey Rajini, I'm not too familiar with that process or potentially the greater solution you're trying to build here. I'd recommend reaching out to support with an incident requesting more details or posting to the Birst ThinkTank community where they might be more readily familiar with what you're looking to do :)RE: [How To] Load data from a database to the Data Lake by Deepak Jadhav:
will i be able to attach M3 database using anySQL modeler type?
RE: [How To] Load data from a database to the Data Lake by Mike Kalinowski:
You can connect to any databases that you have hosted on-prem or in a private cloud so as long as you’re talking about M3 on-prem, that’s fine.
For M3 in MT, they actually push data and provide a replication framework within the application so AnySQL wouldn’t be available in this particular scenario.
hope that helps!