Querying GraphicsLayer

Discussion created by sohnem2 on Dec 4, 2013
I am haivng to use a graphicslayer instead of a featureslayer because featureslayer does not allow geometry to be applied to it.  So was adding and removing graphics manually to the graphicslayer beased on a query that takes geometry.

Could this be inefficient in terms of lots of graphics?
What are the issues with using graphicslayer as opposed to a featureslayer in terms of performance if we're talking about thousands/tens of thousands of features?

What is the best approach if dealing with so many featuers, and if geometry has to be applied to the queries?
What are optimization techniques one should be aware of in terms of either graphicslayer or featureslayer?
Is there clustering available for Javascript API?