parallel Directory Reference

Files
	comm.cpp
	Process-level communication.
	comm.h
	Process-level communication: MPI when built with Mallard_ENABLE_MPI, otherwise a single-rank stub with the same interface.
	device_comm.cpp
	Stream-ordered communication of device buffers (NCCL).
	device_comm.h
	Stream-ordered communication of device buffers (NCCL), and ordering between execution space instances without host synchronization.
	distributed_mesh.cpp
	Scalable setup of a distributed mesh.
	distributed_mesh.h
	Scalable setup of a distributed mesh: no rank ever holds the whole mesh.
	distribution.cpp
	A rank's share of a distributed mesh.
	distribution.h
	A rank's share of a distributed mesh: owned cells, halo layers and the halo exchange plan.
	global_order.cpp
	Moving per-item values to contiguous blocks in global-id order.
	global_order.h
	Moving per-item values from their owner ranks to contiguous blocks in global-id order, for writes at global offsets.
	halo_exchange.cpp
	Fills halo cells with their owners' values.
	halo_exchange.h
	Fills halo cells with their owners' values.
	partition.cpp
	Cell partitioning for distributed runs.
	partition.h
	Cell partitioning for distributed runs.