dlbtiming.h File Reference

#include <memory>
#include "gromacs/mdtypes/commrec.h"

Include dependency graph for dlbtiming.h:

Description

This file declares functions for timing the load imbalance due to domain decomposition.

Author

Berk Hess hess@.nosp@m.kth..nosp@m.se

Classes
class	DDBalanceRegionHandler
	Manager for starting and stopping the dynamic load balancing region. More...
class	BalanceRegion
	Object that describes a DLB balancing region. More...

Enumerations
enum	DdAllowBalanceRegionReopen { DdAllowBalanceRegionReopen::no, DdAllowBalanceRegionReopen::yes }
	Tells if we should open the balancing region. More...
enum	DdBalanceRegionWaitedForGpu { DdBalanceRegionWaitedForGpu::no, DdBalanceRegionWaitedForGpu::yes }
	Tells if we had to wait for a GPU to finish computation. More...

Functions
void	ddReopenBalanceRegionCpu (const gmx_domdec_t *dd)
	Re-open the, already opened, load balance timing region. More...
void	dd_force_flop_start (struct gmx_domdec_t *dd, t_nrnb *nrnb)
	Start the force flop count.
void	dd_force_flop_stop (struct gmx_domdec_t *dd, t_nrnb *nrnb)
	Stop the force flop count.
void	clear_dd_cycle_counts (gmx_domdec_t *dd)
	Clear the cycle counts used for tuning.

Enumeration Type Documentation

enum DdAllowBalanceRegionReopen

strong

Tells if we should open the balancing region.

Enumerator
no	Do not allow opening an already open region.
yes	Allow opening an already open region.

enum DdBalanceRegionWaitedForGpu

strong

Tells if we had to wait for a GPU to finish computation.

Enumerator
no	The GPU finished computation before the CPU needed the result.
yes	We had to wait for the GPU to finish computation.

Function Documentation

void ddReopenBalanceRegionCpu

(

const gmx_domdec_t *

dd

)

Re-open the, already opened, load balance timing region.

This function should be called after every MPI communication that occurs in the main MD loop. Note that the current setup assumes that all MPI communication acts like a global barrier. But if some ranks don't participate in communication or if some ranks communicate faster with neighbors than others, the obtained timings might not accurately reflect the computation time.

Parameters

[in,out]

dd

The domain decomposition struct