The Diagnostics node in the data storage tree offers information about the system load by the data stores. The measurement must be running.

The performance values of all data stores are shown in the table. There is one row per data store. The rows are grouped according to the threads that write the data.

In each group row is the name of the thread and (in brackets) the time utilization (100% means the thread is overloaded). The load average is displayed by default. But you can switch between the average and actual value using the context menu.

The Destination column indicates the respective target to which the data is written, for example a hard disk partition, the address of the database, the address of the Kafka cluster, etc. The Write speed indicates how fast the data is written.

The Memory buffer (kB) columns indicate how much data is buffered in ibaPDA. The columns File buffer (MB) indicate how much data is buffered in the file buffer.

The Acquisition Thread load column indicates various information depending on the data stores. For timebased data stores, the Acquisition Thread load column indicates the amount of time needed for the run length encoding and writing to a disk. For DB/Cloud, MQTT, Kafka Cluster, InfluxDB and MindSphere data stores, the column indicates the load caused by the analysis of the triggers and creation of the row data.

For HD data stores, the partial processing time is displayed, that is used for the creation of the data to be written on the HD server. These values already contain the run length encoding for timebased stores, event trigger calculation for event stores and the calculation of the lengthbased data for lengthbased stores.

Additional information about diagnostics can be found in the ibaPDA manual, part 5.