Run Flow Dialog

Settings

Parameters

Parameters to pass to the flow.

Column	Description
Set	Check to pass a parameter value to the flow. Uncheck to have the flow use the default value for this parameter.
Type	The type of the parameter, as declared by the flow.
Name	The name of the parameter, as declared by the flow.
Value	The value to pass for this parameter.
Description	The description of the parameter, as declared by the flow.

Logging

The log level for the job:

• TRACE - the most verbose logging mode
• DEBUG - much internal activity is logged
• INFO - the logging mode for normal operation
• WARN - only warnings and errors are logged
• ERROR - only errors are logged
• OFF - logging is disabled

The default row logging interval.

Steps generate a progress feedback line in the log whenever they finish processing this many rows.

Memory

Option	Description
Initial Memory	Initial heap memory for the JVM running the job Leave blank or set to 0 to auto-manage memory.
Maximum Memory	Limit for heap memory for the JVM running the job Leave blank or set to 0 to auto-manage memory.

Tracking

Tracking options determine how much information is collected from the running job and displayed in the user interface.

Track

Flows can in turn launch sub-flows. This setting determines which flows to track.

• main flow only
  Runtime information is collected for the main flow only. Any sub-flows are not tracked.

• all flows
  Runtime information is collected for the main flow and any sub-flows it launches. Please note that if the job launches a substantial number of flows, collecting too much information can cause the user interface to become unresponsive.

• disabled
  Does not collect any runtime information for any flows. Only high level information about the run process is available in the user interface.

Capture

This setting determines how processed data is collected for display in the user interface.

Select if and how many data rows to capture from hops, and which flows to collect data from.

Field Size Limit

This limit prevents overloading the data viewing UI with excessively large values.

Determines the maximum size in bytes of any single captured field in a row. Fields bigger than the specified limit are truncated.

• Smaller values increase UI responsiveness, but less field data is preserved for viewing.
• Bigger values show a more complete view of captured data but may impact UI responsiveness.

Speed

This setting allows balancing processing speed vs. responsiveness of the user interface.

A data processing job can easily take up most of the computing resources of a desktop computer. This setting allows to throttle the processing, so the user interface has a chance to catch up and responsively display processed data.

Name

The name of the job.

This name appears on the jobs and logs panel. Distinct names make it easier to distinguish individual runs of the same flow.

Leave empty to use an auto-generated name.

Row Passing

Option	Description
Hop Buffer Size	Buffer size for hops Each hop uses three equally sized buffers used to efficiently exchange rows between the steps it connects. This parameter controls the size of these buffers in rows. Low buffer sizes decrease memory consumption, increase CPU cache utilization, but also increase total CPU row-passing overhead. High buffer sizes increase memory consumption, decrease CPU cache utilization, but also decrease CPU row-passing overhead.
Idle Time	Maximum Worker Idle Time When a step is starved for input or blocked on output, its thread is descheduled to continue processing at a later time. This controls an optimistic wait time in ms during which a worker thread will merely yield and recheck conditions before actually being descheduled. Usually 0 is the most efficient value, causing starved steps to immediately give up processing time and deschedule their threads. This setting is only useful in rare circumstances where pathologic row-dripping timing can cause an otherwise short and efficient pipeline to keep de-scheduling and re-scheduleing in a thrashing manner.