DataFlux Data Management Studio 2.6: User Guide

Displaying Run-Time Statistics in SAS Job Monitor

Overview of SAS Job Monitor

If your site has licensed SAS^® Environment Manager and SAS^® Job Monitor, then you can use a web browser to display run-time statistics for DataFlux Data Management Studio jobs and the nodes within those jobs.

SAS Environment Manager is a web-based monitoring solution for a SAS environment. SAS Job Monitor is an optional component of SAS Environment Manager. It reads job logs at specified locations and displays run-time statistics from the logs. If you can access the SAS Environment Manager, then you can select Analyze > SAS Job Monitor to display run-time statistics for DataFlux Data Management Studio data jobs and process jobs.

Prerequisites for Run-Time Statistics in SAS Job Monitor

You cannot use SAS Job Monitor to display run-time statistics for jobs that are executed interactively in DataFlux Data Management Studio. DataFlux Data Management Studio does not generate a unique log for each job.

The jobs must be executed with the dmpexec command or with a DataFlux Data Management Server. For more information about the dmpexec command, see Run a Job Whose Run-Time Statistics Will Be Displayed in SAS Job Monitor. For more information about the server, see Deploying Jobs for Execution on a Data Management Server.

Usage Notes for Run-Time Statistics in SAS Job Monitor

By default, job node statistics are logged once per job run: when the job has finished executing. If you want to log node statistics while the job is running, then you must change the value for the BASE/MONITOR_FREQUENCY option in the app.cfg file. For more information about this option, see "Configuration Options" in the DataFlux Data Management Studio Installation and Configuration Guide.

If you update the BASE/MONITOR_FREQUENCY option so that performance statistics are taken a regular intervals, be aware that different nodes might be executing during different intervals. The log will contain statistics only for those nodes that executed during a given interval.

The log does not capture run-time statistics for embedded process job nodes. It captures run-time statistics for embedded data job nodes.

The log will not reliably capture the number of times a node was executed in a loop. This is because the iteration rate might be faster than the interval between snapshots, as specified in the BASE/MONITOR_FREQUENCY option.