We’re trying to load huge CZML files (30mb and more), that come from our server.
Cesium is taking a long time processing the data (up to 1 minute).
Already tried suspending/resuming events, but there were no improvements. I used the Chrome profiler and seems most time is spent on computeAvailiability() for the CZML content.
Any tip/idea how to improve this situation?
Not sure if this will help, but I’m wondering if there are any optimizations when generating the CZML file you’ve tried already? Possibly having the server presort all the entities by their time intervals to reduce the work necessary for the computeAvailability method?
Sorting won’t help here. It’s very odd that computeAvailability is where the profiler is spending its time. What kind of data are you sending over? Can you share one of these 30 meg files (even if only privately). If so, I can take a quick look.
Do you actually specify availability for the entities?