html_url,issue_url,id,node_id,user,created_at,updated_at,author_association,body,reactions,performed_via_github_app,issue
https://github.com/pydata/xarray/issues/2504#issuecomment-432735579,https://api.github.com/repos/pydata/xarray/issues/2504,432735579,MDEyOklzc3VlQ29tbWVudDQzMjczNTU3OQ==,1217238,2018-10-24T16:44:23Z,2018-10-24T16:44:23Z,MEMBER,"I think this is basically a netCDF4-Python issue. It is definitely a known issue that some complex indexing operations can be more expensive when done on netCDF4-Python arrays rather than in-memory NumPy arrays.

The easy work-around is to write something like `ds.load()` first to load the data into memory. Unfortunately this is hard to fix in general because there is a tradeoff in terms of memory usage. 

The other option is to try another netCDF4 backend engine, e.g., install h5netcdf and set `engine='h5netcdf'`.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,373449569