In this video I’ll be answering the question, what is the PI Interface for RDBMS? I’ll give some examples where you’d want to use this PI Interface. I’ll explain the data source requirements and finally I’ll go over the steps for how to start collecting data. So what’s the PI Interface for RDBMS? It’s a PI Interface that collects data from Relational Database Management Systems. Common examples include the Microsoft SQL Server, Oracle database and IBM DB2. The full name is actually the PI Interface for Relational Database RDBMS via ODBC meaning it can collect data from any RDBMS that has an ODBC driver. Some of you might be thinking RDBMS, ODBC, what is all of this stuff? Let’s start with what is a relational database? First a database is just a collection of data that is organized. A relational database is data that’s organized in tables which have relationships. Let’s see an example. Say we have a database that contains manufacturing data, the database contains a table called Material Lots which tracks the batches of products produced. It also has a table called inventory which tracks where the batches are stored and products with all the different product numbers and names. As we can see, the columns in one table often correspond to a column in another table. So these tables have relationships that link them, making this a relational database. A relational database management system or RDBMS is simply the software that you use to create and manage this type of database. So what kind of data would you want to collect from a relational database? We know that PI Interfaces collect time series data and send it to the PI Data Archive. So really we could collect any data from a RDBMS where the time is important. Here are some common examples. You can collect data from a manufacturing execution system. For example you could collect the time where specific batches of product were produced like the batches of dairy products we saw above. You could collect data from a laboratory information management system such as the results of lab tests related to your process. You can collect data from an enterprise resource management system such as financial data related to your product at any given time. And finally although less common, you can also use the PI Interface for RDBMS to write data from a PI Point back to an RDBMS. For example you could write production totals back into your manufacturing database. So how does this single PI Interface collect data from all of these systems? As we mentioned the PI Interface can collect data from any RDBMS with an OBDC driver. OBDC stands for Open Database Connectivity. This is simply a standard way of communicating with Database Management Systems. An OBDC driver is just a software component that does the talking. This driver needs to be installed on the computer where you’ll be installing the PI Interface for RDBMS. Alright, now that you know the basics, what are the steps for setting up a PI Interface for RDBMS? First you’ll want to install the PI Interface for RDBMS and the PI Interface Configuration Utility. We recommend that you install these components on a dedicated company, so not on the PI Data Archive or on the RDBMS Server. Then you’ll want to configure and run an instance of the PI Interface and configure buffering. Then you’ll need to choose a data retrieval strategy for your PI Points. You can use the single tag strategy, tag group strategy, or tag distribution strategy. Next you’ll need to build your SQL queries. SQL stands for Structured Query Language and is the language that PI Interfaces use to ask the RDBMS for data. The final step is to create your PI Points. Each of these topics will be covered in the videos in this playlist. One final note is that throughout the rest of the playlist, the term PI Server will be used to refer to the PI Data Archive.