Due to the current gap in continued funding from the U.S. National Science Foundation (NSF), the NSF Unidata Program Center has temporarily paused most operations. See NSF Unidata Pause in Most Operations for details.
Tim Hume wrote: > I was wondering about how efficient the NetCDF reading and writing > functions are. In particular, it should be possible to use the NetCDF > functions: > > nc_get_varm_type and > nc_put_varm_type > > to read and write any arbitrary array of NetCDF data. My question is, is > it less efficient to use the put/get_varm functions than using one of the > more "specialised" functions (put_var1, put_var, put_vars, put_vara etc)? In the C interface, the vars (strided) interfaces are implemented in terms of the varm (mapped) interfaces, which are implemented using the vara (array) interfaces. The mapped interfaces don't offer any significant efficiency advantages over the more specialized interfaces, but may offer convenience if they do what you want. They have not been widely used, and were not implemented in the newer object-oriented interfaces, such as for Java and C++. --Russ _____________________________________________________________________ Russ Rew UCAR Unidata Program russ@xxxxxxxxxxxxxxxx http://www.unidata.ucar.edu
netcdfgroup
archives: