5 Things to Know About Importing Data into IBM Intelligent Operations Center
MarcelaAdanRochester 2700048UPS Visits (8171)
Bringing data into the IBM Intelligent Operations Center solution is a critical process. The information may come in different formats and from different kinds of sources. The requirements on how to import the data also vary. It is very important to understand the key considerations that should be taken into account when planning to connect to a new data source to import data into IBM Intelligent Operations Center. Here are five things to consider when you are planning importing data into IBM Intelligent Operations Center:
1. The data items in IBM Intelligent Operations Center are flat.
The data in IBM Intelligent Operations Center is flat: each data item is represented as a row in a table. If the original data has hierarchical properties, or has properties stored in multiple database tables or even in multiple systems, the data needs to be transformed and/or combined before configuring the standard IBM Intelligent Operations Center data source. The workaround for hierarchical data is to use extension properties as described in the IBM Redbooks publication IBM
For the data that has properties stored in multiple database tables, DB2 views can be used as an alternative way of representing data that exists in one or more tables. If the data to be imported is stored in multiple systems, then there are two IBM Intelligent Operations Center components that can help to combine and transform the data before importing it:
2. Data items properties formats
The minimal data properties for location and time stamps must be transformed to the following formats:
3. Store binary data item properties in external systems.
If the use case requires to bring binary data into IBM Intelligent Operations Center to be displayed in the user interface, for example, image, audio, or video files, the binary data must be hosted in an external system (for example, an HTTP server for images). Then, render the binary data on a preview card with custom code as described in the book IBM
4. Decide whether the data items can be pulled into IBM Intelligent Operations Center at regular intervals or they must be pushed immediately.
If the data items can be pulled into IBM Intelligent Operations Center based on a scheduled batch upload, then CSV or database data source types are a good choice. If the data items must be pushed into IBM Intelligent Operations Center as soon as they are generated, then you should use the data injection service, via either the Java or the REST APIs provided with IBM Intelligent Operations Center or a CAP data source.
5. Determine the value of the data before importing it into IBM Intelligent Operations Center.
There are some use cases where data items are generated by sensors or other types of external systems at a high speed, velocity, and variety. Not all this data might be required in IBM Intelligent Operations Center and some of it may be of no use to the end user. For example, it is more useful for the operator to get an alert when a particular threshold is reached by one of the sensors, rather than getting messages on every value that changed. It is a good practice to filter data items before sending them into the system. Another option is to create a new record in the staging database table or CSV file using analytics tools such as IBM InfoSphere Streams, which can analyze the data streams, transform, and filter them before sending an event or notification to IBM Intelligent Operations Center. You can also filter data with routing rules, by routing to IBM Intelligent Operations Center only the data items that meet a pre-defined criteria. This feature is especially useful when the data is coming in CAP format and it allows to route different CAP messages to different data sources. For more information see IBM
For more information on integrating IBM Intelligent Operations Center with external data sources, refer to the IBM Redbooks publication IBM
The following tutorials demonstrate how to configure IBM Intelligent Operations Center data sources:
See also the blog post Unde
Nikolay Vlasov is a Technical Sales Specialist in IBM Singapore. Nikolay helps clients in the ASEAN region to identify their business requirements and design solutions based on IBM Intelligent Operations Center and other products in the IBM Smarter Cities portfolio. Nikolay's previous assignments include technical sales for IBM WebSphere Portal and IBM Collaboration solutions in Russia and CIS countries. Nikolay holds a Master’s degree in Computer Science from Moscow State University, Russia. Nikolay is a co-author of the IBM Redbooks publications IBM Intelligent Operations Center 1.6 Programming Guide, SG24-8201.