This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.
Giang, > To: address@hidden > From: Giang Nong <address@hidden> > Subject: read ahead and write behind on NetCDF > Organization: GFDL/NOAA > Keywords: 200405242114.i4OLENtK012695 netCDF The above message contained the following: > I use IRIX 6.5 and I want to try using netcdf_ffiospec. I want to use > asynchronous read/write and I try the following setting: > netcdf_ffiospec = cachea:256:8:2 > > My question: is it enough to specify netcdf_ffiospec to get asynchronous I/O, > in other words, do I need to get into source code of NetCDF and make some > change there. In case I need to modify NetCDF source code, what changes are > necessary to get asynchronous I/O? I'm afraid that you can't use the FFIO subsystem on a computer running IRIX because the FFIO subsystem is a feature of Cray (and Cray-like) computers only. The "ffio" module of the netCDF package was developed to allow netCDF users on Cray systems to use the Cray FFIO subsystem when accessing netCDF files. The only other I/O subsystem within the netCDF package is the "posixio" module. This module uses simple POSIX read() and write() calls and doesn't support asynchronous I/O. Doing what you want would require the writing of a new I/O subsystem for the netCDF package. Regrettably, we don't have the resources to do that at this time. > Thank you very much. > > Giang Nong > GFDL/NOAA Regards, Steve Emmerson > NOTE: All email exchanges with Unidata User Support are recorded in the > Unidata inquiry tracking system and then made publically available > through the web. If you do not want to have your interactions made > available in this way, you must let us know in each email you send to us.