Replying to [comment:2 scott.gonzalez]:
This proxy has nothing to do with working around any bugs. This is the implementation of the remove event for auto-destruction of widgets. The comment with the bug URL is pointing to the reason that they try/catch exists. The same try/catch was added in jQuery core 1.6.3. I think you're confused about what's going on here. It sounds like you're probably doing something specific beyond just making an ajax request. Can you please provide a reduced test case showing the performance problem you're having?
Ok, i (now) understand that the cleanData-Method has been proxied for other reasons than the bug mentioned.
Looking at it now, i'm not sure what may be the reason for it. It can se a significant difference even when the dom is small (percentagevise).
How i got here? I used dynatrace Ajax-edition to analyze the requests and the cleanData-Method showed up very high in the hotspots view. I looked into the jQuery-UI code and found the proxy. Removed it and found client-processing was 60% faster.
I will check the code tomorrow to see if we have some expensive remove-handlers that make this problem specific to our code. It has to be the additional iteration itself or the handlers.
What we are doing? Wicket6-Applications (jQuery) with ajax. In some very large forms the dom can get huge (10 Tabs). But as said, i can see a great difference in small dom-trees as well.
Thank you for your time.