Quick and Dirty Tricks for Large Assembly Management

[Tmetcalf]

There are good bottom-up techniques to build manageable large assemblies from the get-go, but what do you do when  handed a task and a dataset from a less-than-diligent modeler? How do you square away a problem dataset without remodeling the whole thing? What if it’s gotten so large, you can’t even open it?  In this session, the presenter shared some outside-the-box tricks for manipulating large datasets, getting them back into shape and efficiently usable. The presenter touched on config.pro options and standard techniques every large assembly modeler should employ. Attendees learned about creative uses of the model tree, layer management, feature redefinition, trail file scrutiny, and vendor model reconfiguration. And how these techniques  allow modelers to speed up a dataset’s retrieval and manipulation, and get productivity back on track.

[mgebhardt]

In the presentation there is an acronym called "LAMON" - I can derive 'Large Assembly Management' - but out of curiosity what does the "O.N." stand for? .

[Tmetcalf]

Hi Mark -

I will reach out the presenter.

Best,

Toby

[phollmer]

It stands for Large Assembly Management - On. We have another mapkey called "lamoff", which toggles all the settings back to our default configuration.

[mgebhardt]

Thanks Peter!  I'm a TSAM with PTC (Technical Support Account Manager). I'm helping a customer organize their use-cases while they are working through Windchill performance concerns they are having, i.e. "Saving single objects to Large Work-spaces (15K+)".  One of my technical experts in my department is recommending they deploy some 'Large Assembly Management' Practices. Your presentation is awesome so I am hoping the Windchill team at this company can learn from it - and if it works out, be able to attend a breakout on this topic, possibly at the 2017 LiveWorx next year!  Much appreciation! ~ Mark

[phollmer]

No problem! Looking forward to the breakout at 2017. One thing I'd advocate: test out the config.pro settings in your environment. or rather, build the LAM config.pro in small chunks and get feedback on whether things are improving (or measure performance, if you can). Depending on your company's network and other factors, some recommendations may benefit you, other recommendations may actually slow things down, so use what works and discard the rest (and write it down somewhere, when someone asks you two years from now, "why aren't we using X setting?"). Good luck, -Pete