Currently, I'm having a problem with a drawing of a size of approximately 100-200 MB. Loading this drawing and all parts can take up to 10 minutes on a quad-core Xeon machine with 2 threads per machine and ample memory. The funny part however, is that for the loading and processing of the drawing (Up untill the point where I can actually work with it), the machine utilizes only one thread. Is there a way to employ the remaining core threads to speed up this proces? Or is it a physical impossibility, considering that all threads would have to reference the same data model that is being constructed in the memory?