Source code location affects execution speed?!?

[John Lokanis]

This is a very strange problem.  I have put together a demo project to demonstrate the speed differences in array vs variant attribute storage methods.  I also included a test of different variant database designs to show some pitfalls.

I first built this project in the following folder on my machine:

C:DevelopmentSourcesandboxVariant Attributes

It seemed to be working well and demonstrating the effects I was after.

I then moved the folder to my desktop:

C:UserslokanisDesktopVariant Attributes

When I opened the project from here, and tested it, the demo ran 1000x faster and no longer demonstrated the effect I was after.

Next I tried moving it to here:

C:tempVariant Attributes

And I got a result closer to the original location but still not the same.

 

Here is the code: Variant Attributes.zip

 

Run the VI: "Compare Class Database Storage Methods.vi"

 

Any ideas what is going on?

 

BTW: the point of the demo is to show that storing large elements in a variant database is faster if you store DVRs of the elements.  This should avoid large data copies.

 

 

[Ulf]

I can't open or test your example as I'm using LabVIEW 11, but it might be some kind of virus scanning of your file, that affect the execution speed?

[gleichman]

I ran it from my desktop and a directory off the root and they both gave similar results (~330x). LV2013 and Win7sp1. Did you run it both times from inside the project?

[John Lokanis]

Yes, always from inside the project.  I cannot reproduce the 330x result.  This is what I get running outside the project from the desktop:

 

Of course, now I get the same results from all locations.  Must have been something weird about my machine at the time.

 

(Also running Win7 and LV2013)

[shoneill]

Ooh this sounds familiar.

 

I had some WEIRD performance issues a few weeks back with the divide primitive taking different amounts of time depending on which values it was dividing, and I mean MEASURABLE differences.

 

Closed the project, re-opened and the issue was completely gone (hopefully) never to be seen again.

 

It seemed at the time like the math libraries had started going a bit wonky.

 

As with all problems of this ethereal type (you look at them properly and they disappear) you end up questioning your sanity and have no change of filing a bug report because noone will be able to reproduce it.

 

Shane.

[Mellroth]

I also had a similar issue where a simple VI using Variant attribute lookup took a long time to execute if the owning library/class was opened.

The time difference was never measured, but running through ~1000 elements was noticeable slower.

 

I tracked this down to be an issue with re-entrancy, once I made the VIs reentrant the execution speed was the same regardless of where it was started.

 

/Jonas