Best practice to handle heavy elevation point data,
I couldn’t figure out if Terrain dataset or LAS dataset is better option to handle huge elevation point data (billion of points)
I Think before you even ask what is the beat way to handle the point cloud you need to understand what you want to use the data for. Large point clouds can be managed through LAS datasets, Mosaic datasets or geodatabase terrain. The former two preserve the point cloud while the last is a filtered representation of the original data.
if you need to get access to all classifications of the point cloud and need to perform feature extraction the the LAS dataset is the way to go.
if you want to interpret the surface and have access to the data as a seamless surface them mosaic dataset, feeding an elevation service is the way to go.
if you no not need to preserve the underlying data, want to interpret just a DEM or DSM and incorporate other data sources into the terrain then a geodatabase terrain can be useful.
if you know what you need to do with the data the the best management tool then becomes clear.
Lidar and other point clouds
Imagery and Remote Sensing
source of discussion
Using lidar in ArcGIS—Help | ArcGIS for Desktop
help topic covering lidar data its use in several situations (las and terrain)
Advantages of using lidar in GIS—Help | ArcGIS for Desktop
Benefits of using LAS datasets—Help | ArcGIS for Desktop
What is a terrain dataset?—Help | ArcGIS for Desktop (requires 3D)
as a start
Many thanks Dan and Gordon for the input. It really helps.
I would second the above posts by Dan and Gordon, and also probably say that if you're using ArcGIS for your file management, then it's best to use LAS Dataset rather than any kind of tool that manages a raster elevation derivative. The reason for this is that you will always have your raw data handy to filter, edit classifications, or make other derivatives whenever you want, and you can add/remove tiles as necessary. Just my preference, though.
Retrieving data ...