Drivers

evolvers

drivers

Tasks are small groups of steps that can be completed without adversely affecting user interface responsiveness. Stages are larger units specified by the MIF problem description; in particular, problem parameters are not expected to change in a discontinuous manner inside a stage. The run is the complete sequence of stages, from problem start to finish. The driver detects when stages and runs are finished, using criteria specified in the MIF problem description, and can enforce constraints, such as making sure stage boundaries respect time stopping criteria.

There are two drivers in Oxs, Oxs_TimeDriver for controlling LLG time evolvers such as Oxs_RungeKuttaEvolve, and Oxs_MinDriver for controlling energy minimization evolvers like Oxs_CGEvolve.

Oxs_TimeDriver:

The Oxs time driver is Oxs_TimeDriver. The specify block has the form

Specify Oxs_TimeDriver:name {
evolver evolver_spec
mesh mesh_spec
Ms scalar_field_spec
m0 vector_field_spec
stopping_dm_dt torque_criteria
stopping_time time_criteria
stage_iteration_limit stage_iteration_count
total_iteration_limit total_iteration_count
stage_count number_of_stages
stage_count_check test
checkpoint_file restart_file_name
checkpoint_interval checkpoint_minutes
checkpoint_disposal cleanup_behavior
start_iteration iteration
start_stage stage
start_stage_iteration stage_iteration
start_stage_start_time stage_time
start_stage_elapsed_time stage_elapsed_time
start_last_timestep timestep
normalize_aveM_output aveMflag
report_max_spin_angle report_angle
report_wall_time report_time
}

The first four parameters, evolver, mesh, Ms and m0 provide references to a time evolver, a mesh, a scalar field and a vector field, respectively. Here Ms is the pointwise saturation magnetization in A/m, and m0 is the initial configuration for the magnetization unit spins, i.e., $\vert\textbf{m}\vert=1$ at each point. These four parameters are required.

The next group of three parameters control stage stopping criteria. Measured in degrees per nanosecond, the stopping_dm_dt value specifies that a stage should be considered complete when the maximum $\vert d\textbf{m}/dt\vert$ across all spins drops below this value. Similarly, the stopping_time value specifies the maximum “Simulation time,” i.e., the Landau-Lifshitz-Gilbert ODE (7.2), (7.3) time, allowed per stage. For example, if time_criteria is 1e-9, then no stage will evolve for more than 1 ns. If there were a total of 5 stages in the simulation, then the total simulation time would be not more than 5 ns. The third way to terminate a stage is with a stage_iteration_limit. This is a limit on the number of successful evolver steps allowed per stage. A stage is considered complete when any one of these three criteria are met. Each of the criteria may be either a single value, which is applied to every stage, or else a grouped list of values. If the simulation has more stages than a criteria list has entries, then the last criteria value is applied to all additional stages. These stopping criteria all provide a default value of 0, meaning no constraint, but usually at least one is specified since otherwise there is no automatic stage termination control. For quasi-static simulations, a stopping_dm_dt value in the range of 1.0 to 0.01 is reasonable; the numerical precision of the energy calculations usually makes it not possible to obtain $\vert d\textbf{m}/dt\vert$ much below 0.001 degree per nanosecond.

total_iteration_limit, stage_count and stage_count_check parameters involve simulation run completion conditions. The default value for the first is 0, interpreted as no limit, but one may limit the total number of steps performed in a simulation by specifying a positive integer value here. The more usual run completion condition is based on the stage count. If a positive integer value is specified for stage_count, then the run will be considered complete when the stage count reaches that value. If stage_count is not specified, or is given the value 0, then the effective number_of_stages value is computed by examining the length of the stopping criteria lists, and also any other Oxs_Ext object that has stage length expectations, such as Oxs_UZeeman. The longest of these is taken to be the stage limit value. Typically these lengths, along with stage_count if specified, will all be the same, and any differences indicate an error in the MIF file. Oxs will automatically test this condition, provided stage_count_check is set to 1, which is the default value. Stage length requests of 0 or 1 are ignored in this test, since those lengths are commonly used to represent sequences of arbitrary length. At times a short sequence is intentionally specified that is meant to be implicitly extended to match the full simulation stage length. In this case, the stage count check can be disabled by setting test to 0.

The options are used to control the saving of solver state to disk; these saves are used by the oxsii and boxsi restart feature. The value of the checkpoint_file option is the name to use for the solver state file. The default is base_file_name.restart.

Cleanup of the checkpoint file is determined by the setting of checkpoint_disposal, which should be one of standard (the default), done_only, or never. Under the standard setting, the checkpoint file is automatically deleted upon normal program termination, either because the solver reached the end of the problem, or because the user interactively terminated the problem prematurely. If cleanup_behavior is set to done_only, then the checkpoint file is only deleted if the problem endpoint is reached. If cleanup_behavior is never, then OOMMF does not delete checkpoint file; the user is responsible for deleting this file as she desires.

The checkpoint_interval value is the time in minutes between overwrites of the checkpoint file. No checkpoint file is written until checkpoint_minutes have elapsed. Checkpoint writes occur between solver iterations, so the actual interval time may be somewhat longer than the specified time. If checkpoint_minutes is 0, then each step is saved. Setting checkpoint_minutes to -1 disables checkpointing. The default checkpoint interval is 15 minutes.

The six start_* options control the problem run start point. These are intended primarily for automatic use by the restart feature. The default value for each is 0.

The normalize_aveM_output option is used to control the scaling and units on the average magnetization components , and sent as DataTable output (this includes output sent to mmDataTable, mmGraph, and mmArchive). If aveMflag is true (1), then the output values are scaled to lie in the range , where the extreme values are obtained only at saturation (i.e., all the spins are aligned). If aveMflag is false (0), then the output is in A/m. The default setting is 1.

In the older MIF 2.1 format, the driver Specify block supports three additional values: basename, scalar_output_format, and vector_field_output_format. In the MIF 2.2 format these output controls have been moved into the SetOptions block. See the SetOptions

documentation for details.

Oxs_TimeDriver provides 12 scalar outputs and 2 vector field outputs. The scalar outputs are

Stage: current stage number, counting from 0.
Stage iteration: number of successful evolver steps in the current stage.
Iteration: number of successful evolver steps in the current simulation.
Simulation time: Landau-Lifshitz-Gilbert evolution time, in seconds.
Last time step: The size of the preceding time step, in seconds.
Mx/mx: magnetization component in the direction, averaged across the entire simulation, in A/m (Mx) or normalized units (mx), depending on the setting of the normalize_aveM_output option.
My/my: magnetization component in the direction, averaged across the entire simulation, in A/m (My) or normalized units (my), depending on the setting of the normalize_aveM_output option.
Mz/mz: magnetization component in the direction, averaged across the entire simulation, in A/m (Mz) or normalized units (mz), depending on the setting of the normalize_aveM_output option.
Max Spin Ang: maximum angle between neighboring spins having non-zero magnetization , measured in degrees. The definition of “neighbor” depends on the mesh, but for Oxs_RectangularMesh the neighborhood of a point consists of 6 points, those nearest forward and backward along each of the 3 coordinate axis directions.
Stage Max Spin Ang: the largest value of “Max Spin Ang” obtained across the current stage, in degrees.
Run Max Spin Ang: the largest value of “Max Spin Ang” obtained across the current run, in degrees.
Wall time: Wall clock time, in seconds.

The three “Max Spin Ang” outputs are disabled by default. In general one should refer instead to the neighboring spin angle outputs provided by the exchange energies. However, for backward compatibility, or for simulations without any exchange energy terms, the driver spin angle outputs can be enabled by setting the report_max_spin_angle option to 1.

The “Wall time” output is also disabled by default. It can be enabled by setting the report_wall_time option to 1. It reports the wall clock time, in seconds, since a system-dependent zero-time. This output may be useful for performance comparisions and debugging. (Note: The timestamp for a magnetization state is recorded when output is first requested for that state; the timestamp is not directly tied to the processing of the state.)

The vector field outputs are

Magnetization: magnetization vector $\textbf{M}$ , in A/m.
Spin: unit magnetization $\textbf{m}$ . This output ignores the vector_field_output_format precision setting, instead always exporting at full precision.

Examples: sample.mif, pulse.mif.

Oxs_MinDriver:

The Oxs driver for controlling minimization evolvers is Oxs_MinDriver. The specify block has the form

Specify Oxs_MinDriver:name {
evolver evolver_spec
mesh mesh_spec
Ms scalar_field_spec
m0 vector_field_spec
stopping_mxHxm torque_criteria
stage_iteration_limit stage_iteration_count
total_iteration_limit total_iteration_count
stage_count number_of_stages
stage_count_check test
checkpoint_file restart_file_name
checkpoint_interval checkpoint_minutes
checkpoint_disposal cleanup_behavior
start_iteration iteration
start_stage stage
start_stage_iteration stage_iteration
start_stage_start_time stage_time
start_stage_elapsed_time stage_elapsed_time
start_last_timestep timestep
normalize_aveM_output aveMflag
report_max_spin_angle report_angle
report_wall_time report_time
}

These parameters are the same as those described for the Oxs_TimeDriver class, except that stopping_mxHxm replaces stopping_dm_dt, and there is no analogue to stopping_time. The value for stopping_mxHxm is in A/m, and may be a grouped list . Choice depends on the particulars of the simulation, but typical values are in the range 10 to 0.1. Limits in the numerical precision of the energy calculations usually makes it not possible to obtain $\vert\textbf{m}\times\textbf{H}\times\textbf{m}\vert$ below about 0.01 A/m. This control can be disabled by setting it to 0.0.

As with Oxs_TimeDriver, in the older MIF 2.1 format this Specify block supports three additional values: basename to control output filenames, and output format controls scalar_output_format and vector_field_output_format. In the MIF 2.2 format these output controls have been moved into the SetOptions block. See the SetOptions

documentation for details.

Oxs_MinDriver provides 10 scalar outputs and 2 vector field outputs. The scalar outputs are

Stage: current stage number, counting from 0.
Stage iteration: number of successful evolver steps in the current stage.
Iteration: number of successful evolver steps in the current simulation.
Mx/mx: magnetization component in the direction, averaged across the entire simulation, in A/m (Mx) or normalized units (mx), depending on the setting of the normalize_aveM_output option.
My/my: magnetization component in the direction, averaged across the entire simulation, in A/m (My) or normalized units (my), depending on the setting of the normalize_aveM_output option.
Mz/mz: magnetization component in the direction, averaged across the entire simulation, in A/m (Mz) or normalized units (mz), depending on the setting of the normalize_aveM_output option.
Max Spin Ang: maximum angle between neighboring spins having non-zero magnetization , measured in degrees. The definition of “neighbor” depends on the mesh, but for Oxs_RectangularMesh the neighborhood of a point consists of 6 points, those nearest forward and backward along each of the 3 coordinate axis directions.
Stage Max Spin Ang: the largest value of “Max Spin Ang” obtained across the current stage, in degrees.
Run Max Spin Ang: the largest value of “Max Spin Ang” obtained across the current run, in degrees.
Wall time: Wall clock time, in seconds.

As is the case for the Oxs_TimeDriver, the three “Max Spin Ang” outputs and “Wall time” are disabled by default. They angle outputs are enabled by setting the report_max_spin_angle option to 1, and the wall time output is enabled by setting the report_wall_time option to 1.

The vector field outputs are

Magnetization: magnetization vector $\textbf{M}$ , in A/m.
Spin: unit magnetization $\textbf{m}$ . This output ignores the vector_field_output_format precision setting, instead always exporting at full precision.

Examples: cgtest.mif, stdprob3.mif.