Tools running parallel processes
This package uses mpi4py and dill to spawn processes and execute them.
You need to have an MPI back-end installed on your machine and add the right path on the $LD_LIBRARY_PATH, so that mpi4py can link to it. You should install mpi4py manually by
$ pip install mpi4py
When everything is set, you can install the mpi_map using:
$ pip install mpi_map
map methods of classes
added export of cwd in order to allow import of local modules
Removed disconnect on stop(). This was causing an unnecessary deadlock.
Reverted change of 1.0.17. The problem with disconnect seems to be related to mpi4py.
Removed disconnect from children. Added free on master.
Allowing for the running of method belonging to an object in the distributed memory.
Advanced management of distributed memory
Fixed a leftover bug on the allocation of memory thorugh function evaluation.
Added barrier synchronization to prevent buffer overloading in heavily asynchronous applications
Added barrier sync in binary communications
Fixed handling of non-tuple outputs in mpi_map_alloc_dmem
Release history Release notifications | RSS feed
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.