I have a pretty simple dashboard that displays open purchase order lines. We generally have at max 200 lines. Scrolling through is choppy, especially using arrow keys for navigation.
The BAQ returns all of the records in less than a second. There are only 20 columns in the data set.
Is there any settings or functionality I can optimize on the dashboard that could help speed this up?
Are there any other views in the dashboard? Like selecting the PO in from one grid, shows that PO’s lines in another. Every time you change the row in the driving grid, the other would need to refresh.
No, it is one BAQ and the users normally use the grouping functionality. But the slow behavior is the same regardless of whether that is used.
How is the filtering done? By the BAQ, a dashboard Tracker input, or a filter on the dashboard grid?
It is written into the BAQ. Since it is only open lines/orders it is on the criteria of the POHeader and PODetail tables.
Any chance you could run a client trace and analyze it through the perf & diagnostics tool? Is it choppy after data is returned? If so, are you running the client locally on your machine or via RDP, Parallels, Citrix, VmWare?
Any fields with excessively long text?
Maybe try making some of the fields not shown, and see if that speeds things up.
Does scrolling through the test results in the BAQ designer show any slugishness?
Do you get the same result when “sliding” the scroll bar? I ask this because the current row stays in focus. Selecting a different row may cause some BPM to fire in the background.
One of the fields was the Notes field and it is an update field on the BAQ. I tried hiding it (and another update-able field) on the BAQ. It helped a little bit, but that one of the purposes of the Dashboard, do easily update the notes field.
I will start looking at that, including BPMs that may be there.
@jgehling I am not sure how to run the client trace or analyzing it. Do you know if any resources I could look at to figure out how to do that>
Here is a great overview video of the tool:
You would need to enable the trace log in the client via tracing options:
In the options ensure “Include server trace is on”
Then in PDT load in the txt file under client trace analysis. Typically rows highlighted in red show potential issues in the client trace.