z/TPF - Group home

Before PJ46737, when using the real-time insights dashboard starter kit to view JVM, CDC, and name-value pair data collected with the ZRTMC command, you could only view data for one z/TPF system at a time. Complex-wide dashboard support (PJ46737) includes new and updated dashboards that you can use to monitor any number of z/TPF systems simultaneously. These complex-wide dashboards also introduce multiple metrics that are useful for non-loosely coupled environments.

The z/TPF real-time insights dashboard starter kit was updated with the following enhancements:


New Dashboards. A new Grafana folder titled ZRTMC Complex-Wide Results was created for four new dashboards to monitor different aspects of your z/TPF systems. As mentioned previously, you can use these dashboards to monitor all your z/TPF systems simultaneously.

1. System State dashboard – Monitor system-level metrics like CPU.
2. Name-value Pair Metrics – Monitor key metrics for any combination of name-value pairs. These combinations of name-value pairs are called application dimensions. More information about creating your own custom application dimensions will be discussed later in this blog post.
3. Business Event Metrics – Monitor rates for business events and dispatch adapters.
4. REST Services – Monitor rates of operations and errors for REST services.

New System-level Metrics. There are eight new metrics available for you z/TPF systems.

1. lp_cpu – The CPU utilization for ECBs marked as low-priority.
2. workload_cpu – The CPU utilization excluding low-priority ECBs.
3. fenced_pseudo_cpu – The pseudo CPU utilization on fenced I-streams.
4. sys_pseudo_cpu – The pseudo CPU utilization on non-fenced I-streams.
5. est_msg_rate – An estimate of the rate of incoming messages. This metric is based on one of your application dimensions.
6. est_msg_exist – An estimate of the average CPU existence time of a message. This metric is based on one of your application dimensions.
7. est_msg_cpu – An estimate of the average CPU that is used during the processing of a message. This metric is based on one of your application dimensions.
8. est_msg_find_file – An estimate of the average number of finds and files that are performed during processing of a message. This metric is based on one of your application dimensions.

JVM Dashboard Updates. JVM monitoring dashboards were updated to view JVM data for any number of z/TPF systems simultaneously.


Define Custom Application Dimensions. As mentioned previously, an application dimension is a combination of name-value pairs. Key metrics are generated for each application dimension to be displayed in the dashboards. For example, analysis of an application dimension Message Type and Code Package would produce key metrics for all unique combinations of the MsgType and IOwnerHi name-value pairs. With this enhancement you can see metrics in the dashboards for your own internal name-value pairs. You can also utilize this enhancement to see metrics for automatically generated name-value pairs, like ISrvcName, without any application code changes. Application dimensions can be defined in the tpf_zrtmc_analyzer_profile.yml. For more information, see the readme file for the z/TPF real-time insights dashboard starter kit.









MySQL Database Support. The real-time insights dashboard starter kit now supports MySQL as a second database option, in addition to MariaDB.


New CDC JSON Data. Two new CDC data types are available in JSON format through the starter kit.

1. Common Deploy Files
2. DASD Pools

For more information about APAR PJ46737, see the APEDIT. For more information about the z/TPF real-time insights dashboard starter kit, see the z/TPF Tools download page.