With IBM Common Data Provider for z Systems, you can stream Z IT operational data to the analytics platform of your choice (including Splunk and the Elastic Stack), which means that you can view IBM Z alongside ops data from other platforms for a complete view of your enterprise.
Two very important aspects for a “data streamer” such as IBM Common Data Provider for z Systems are:
1. To collect the largest variety of data, including specific data source types of interest to the end user.
2. To give the end user the control over the data sent to the analytic platform (consumers).
Common Data Provider supports of a variety of data
The Common Data Provider already supports 140+ data source types, and IBM continuously adds more types. For example the support for CA Top Secret and ACF2 SMF records was added in December 2017, and support for IMS log was added in March 2018.
The Common Data Provider also includes the Open Streaming API, which you can use to send your application data to the Common Data Provider and stream it to analytics platforms.
Now, you can also use the System Data Engine language (a proprietary programming language) of IBM Common Data Provider for z Systems to define new data source types, which are then available in the Configuration Tool to include in your policy. For example, a new data source type can be an SMF record that is not yet supported by the Common Data Provider, or a derivative of an SMF record that is already supported by the Common Data Provider. Users can also apply filters to data source types, and stream the associated data to the analytics platform.
For more information, see the IBM Common Data Provider for z Systems documentation.
Common Data Provider gives the end user control over the data
The end user must have control over the data that is sent to analytics platforms, especially due to cost and data access factors.
The cost of some analytic platforms is based on the quantity of data ingested. For example, the more data that is sent and ingested, the greater the cost. Other aspects of cost are network traffic and resource consumption due to the streaming of data. For example, an enterprise might frequently have millions of SMF records generated each day (or even more in a large environment), and managing the streaming of all these records requires network bandwidth and more resources.
The segregation of data access is very important for many enterprises. An enterprise likely wants to restrict access to specific data to only the part of the enterprise that is implementing a use case that requires the data. Here are some examples:
• An SMF record can contain multiple fields, but not all of the fields are required to implement each use case. Therefore, an enterprise needs the capability to select which fields must be streamed to the analytics platform.
• An SMF record type (such as type 110) contains multiple records, but most likely, the enterprise is interested only in the records that represent a problem or an exception (such as a CICS transaction that runs longer than a certain time, or that abended). Therefore, an enterprise needs the capability to select which records must be streamed to the analytics platform.
By using the System Data Engine language, you can now define which fields of an SMF record, and which records from an SMF record type, to stream to the analytics platform. For filtering records from an SMF record type, you can request records that match a specific string in a field or numeric value.
More information
• For more information about the features, see the IBM Common Data Provider for z Systems documentation and take a look at this blog
• For the latest product news, and information about all IBM products in the IT Service Management area, subscribe to the quarterly ITSM Newsletter.

Join The Discussion

Your email address will not be published. Required fields are marked *