#include <gromacs/taskassignment/taskassignment.h>

Description

Contains the GPU task assignment for all ranks on this physical node.

This can be used to configure the modules that might run tasks on GPUs.

This assignment is made by a GpuTaskAssignmentsBuilder object.

Public Member Functions
	GpuTaskAssignments (GpuTaskAssignments &&source) noexcept=default
	Public move constructor to use with the builder.

void	reportGpuUsage (const MDLogger &mdlog, bool printHostName, bool useGpuForBonded, PmeRunMode pmeRunMode, bool useGpuForUpdate)
	Log a report on how GPUs are being used on the ranks of the physical node of rank 0 of the simulation. More...

void	logPerformanceHints (const MDLogger &mdlog, size_t numCompatibleGpusOnThisNode)
	Logs to `mdlog` information that may help a user learn how to let mdrun make a task assignment that runs faster. More...

DeviceInformation *	initDevice (int *deviceId) const
	Return handle to the initialized GPU to use in the this rank. More...

bool	thisRankHasPmeGpuTask () const
	Return whether this rank has a PME task running on a GPU.

bool	thisRankHasAnyGpuTask () const
	Return whether this rank has any task running on a GPU.

Friends
class	GpuTaskAssignmentsBuilder

Member Function Documentation

DeviceInformation * gmx::GpuTaskAssignments::initDevice ( int * deviceId ) const

Return handle to the initialized GPU to use in the this rank.

Parameters

[out] deviceId Index of the assigned device.

Returns: Device information on the selected devicce. Returns nullptr if no GPU task is assigned to this rank.

void gmx::GpuTaskAssignments::logPerformanceHints	(	const MDLogger &	mdlog,
		size_t	numCompatibleGpusOnThisNode
	)

Logs to mdlog information that may help a user learn how to let mdrun make a task assignment that runs faster.

Parameters

[in]	mdlog	Logging object.
[in]	numCompatibleGpusOnThisNode	The number of compatible GPUs on this node.

void gmx::GpuTaskAssignments::reportGpuUsage	(	const MDLogger &	mdlog,
		bool	printHostName,
		bool	useGpuForBonded,
		PmeRunMode	pmeRunMode,
		bool	useGpuForUpdate
	)

Log a report on how GPUs are being used on the ranks of the physical node of rank 0 of the simulation.

Todo:: It could be useful to report also whether any nodes differed, and in what way.

Parameters

[in]	mdlog	Logging object.
[in]	printHostName	Print the hostname in the usage information.
[in]	useGpuForBonded	Whether GPU PP tasks will do bonded work on the GPU.
[in]	pmeRunMode	Describes the execution of PME tasks.
[in]	useGpuForUpdate	Whether the update is offloaded on the GPU.

Exceptions

std::bad_alloc if out of memory

The documentation for this class was generated from the following files: