MPI¶
Collective Behavior¶
openPMD-api is designed to support both serial as well as parallel I/O. The latter is implemented through the Message Passing Interface (MPI).
A collective operation needs to be executed by all MPI ranks of the MPI communicator that was passed to openPMD::Series
.
Contrarily, independent operations can also be called by a subset of these MPI ranks.
For more information, please see the MPI standard documents, for example MPI-3.1 in “Section 2.4 - Semantic Terms”.
Functionality |
Behavior |
Description |
---|---|---|
|
collective |
open and close |
|
collective |
read and write |
|
independent |
declare and open |
|
collective |
explicit open |
|
independent |
declare, open, write |
|
independent |
declare, open, write |
|
backend-specific |
declare, write |
|
independent |
open, reading |
|
independent |
declare, open, write |
backend-specific |
declare, write |
|
|
backend-specific |
declare, write |
|
independent |
write |
|
independent |
read |
|
collective |
read, immediate result |
- 1(1,2,3,4,5,6)
Individual backends, i.e. parallel HDF5, will only support independent operations if the default, non-collective (aka independent) behavior is kept. Otherwise these operations are collective.
- 2
Dataset declarations in parallel HDF5 are only non-collective if chunking is set to
none
(auto
by default). Otherwise these operations are collective.- 3(1,2)
HDF5 only supports collective attribute definitions/writes; ADIOS1 and ADIOS2 attributes can be written independently. If you want to support all backends equally, treat as a collective operation. Note that openPMD represents constant record components with attributes, thus inheriting this for
::makeConstant
.- 4(1,2)
We usually open iterations delayed on first access. This first access is usually the
flush()
call after astoreChunk
/loadChunk
operation. If the first access is non-collective, an explicit, collectiveIteration::open()
can be used to have the files already open.
Alternatively, iterations might be accessed for the first time by immediate operations such as ::availableChunks()
.
Tip
Just because an operation is independent does not mean it is allowed to be inconsistent.
For example, undefined behavior will occur if ranks pass differing values to ::setAttribute
or try to use differing names to describe the same mesh.
Efficient Parallel I/O Patterns¶
Note
This section is a stub. We will improve it in future versions.
Write as large data set chunks as possible in ::storeChunk
operations.
Read in large, non-overlapping subsets of the stored data (::loadChunk
).
Ideally, read the same chunk extents as were written, e.g. through ParticlePatches
(example to-do).
See the implemented I/O backends for individual tuning options.