2009 Unidata NetCDF Workshop for Developers and Data Providers > Parallel I/O with NetCDF
21.1 Introduction to Parallel I/O
Parallel I/O allows many processes to read/write netCDF data at the
same time.
- Used properly, parallel I/O allows users to overcome I/O
bottlenecks in high performance computing environments.
- A parallel I/O file system is required for much improvement in I/O
throughput.
- NetCDF-4 (and pNetCDF) work with the Message Passing Interface,
version 2 (MPI2). There are many software libraries which support
MPI2. Any supercomputer will have some MPI2 library.
- Other methods of parallelism are not supported by netCDF.
- Parallel I/O can be developed on a Linux workstation, and parallel
performace gains can be seen on multi-core machines. Machines will
hundreds of processors can be simulated.
- For netCDF testing we use
the MPICH2
library.
2009 Unidata NetCDF Workshop for Developers and Data Providers > Parallel I/O with NetCDF