JP7409100B2 - Information processing device, magnetic memory element simulator system, and magnetic memory element simulator method - Google Patents

Description

The present invention relates to an information processing device, a magnetic memory element simulator system, and a magnetic memory element simulator method.

As expectations for spintronic devices increase, it is extremely important to use micromagnetic simulations to calculate the spin torque acting on the magnetic tunnel junction structure of magnetic memory elements. The spin torque is calculated using the magnetization vector of the mesh of the free layer and reference layer that face each other with the barrier layer in between of the magnetic memory element.

When modeling a magnetic memory element, if each layer is divided into triangular prism or hexahedral meshes, the surfaces of the meshes of each layer will overlap exactly at the boundary between the layers. As a result, meshes can be matched one-to-one between layers, making it easier to calculate the spin torque acting on the magnetic tunnel junction structure.

Japanese Patent Application Publication No. 2008-122154 Japanese Patent Application Publication No. 9-266299

However, in the conventional technology, there is a problem in that when each layer of a magnetic memory element is modeled by dividing it into a tetrahedral mesh, it may be difficult to simulate the spin torque acting on the magnetic tunnel junction structure.

For example, consider a magnetic memory element having a structure in which a free layer, a barrier layer, and a reference layer are stacked in this order. At this time, when each layer is divided by a tetrahedral mesh, it may not be possible to make the mesh surfaces of the free layer and the reference layer (for example, 12S and 32S in FIG. 5) correspond one-to-one.

Here, to mesh the model, software called a mesher is used that has a meshing function, but not all meshers have the function of dividing by hexahedral mesh or triangular prism mesh. In particular, non-commercial meshers often do not have a dividing function using hexahedral meshes or triangular prism meshes. On the other hand, almost all meshers, commercial and non-commercial, are equipped with the function of dividing using a tetrahedral mesh. For this reason, for example, a user who is restricted from purchasing or using a mesher that has a dividing function using hexahedral mesh or triangular prism mesh may be able to divide the model only using tetrahedral mesh.

One aspect of the present invention is to simulate spin torque acting on a magnetic tunnel junction structure using a model in which a magnetic memory element is divided into tetrahedral meshes.

In one aspect, the information processing device is characterized by having a reading unit that reads data of a model in which each of the plurality of layers of the magnetic storage element is divided by a tetrahedral mesh. The information processing device is characterized by having a creation unit that creates a triangular prism mesh from a triangular mesh of a boundary surface of any layer when a plurality of layers are stacked, based on data. The information processing device is characterized by having a specifying unit that specifies a tetrahedral mesh included in a triangular prism mesh among the tetrahedral meshes that divide the plurality of layers. The information processing device is characterized in that it has a first calculation unit that calculates the spin torque of the triangular prism mesh by setting the average of the magnetization vectors of the tetrahedral meshes specified by the identification unit as the magnetization vector of the triangular prism mesh. The information processing device is characterized by having a second calculation unit that calculates the spin torque of a tetrahedral mesh included in the triangular prism mesh using the spin torque of the triangular prism mesh calculated by the first calculation unit.

In one aspect, spin torque acting on a magnetic tunnel junction structure can be simulated using a model in which a magnetic memory element is divided into tetrahedral meshes.

FIG. 1 is an explanatory diagram illustrating spin torque. FIG. 2 is an explanatory diagram illustrating mesh surfaces that have one-to-one correspondence. FIG. 3 is a diagram showing an example of a hexahedral mesh. FIG. 4 is a diagram showing an example of a triangular prism mesh. FIG. 5 is an explanatory diagram illustrating superposition of tetrahedral meshes. FIG. 6 is a block diagram showing a configuration example of an information processing device. FIG. 7 is a diagram showing an example of a model divided by a tetrahedral mesh. FIG. 8 is a diagram showing an example of a triangular mesh. FIG. 9 is a diagram showing an example of correspondence data between a virtual triangular prism mesh and a tetrahedral mesh on the free layer side. FIG. 10 is a diagram showing an example of correspondence data between a virtual triangular prism mesh and a tetrahedral mesh on the reference layer side. FIG. 11 is an explanatory diagram illustrating a method for creating a virtual triangular prism mesh. FIG. 12 is a diagram showing an example of a tetrahedral mesh included in a virtual triangular prism mesh. FIG. 13 is a flowchart showing the flow of processing of the information processing device. FIG. 14 is a flowchart showing the flow of processing for associating a virtual triangular prism mesh with a tetrahedral mesh. FIG. 15 is a flowchart showing the flow of processing for calculating the spin torque of a tetrahedral mesh. FIG. 16 is a diagram illustrating an example of a hardware configuration.

Embodiments of an information processing device, a magnetic memory element simulator system, and a magnetic memory element simulator method according to the present invention will be described in detail below with reference to the drawings. Note that the present invention is not limited to this example. Moreover, each embodiment can be combined as appropriate within a consistent range.

The information processing device of this embodiment calculates spin torque from a model of a magnetic tunnel junction (MTJ) structure divided by a tetrahedral mesh. First, spin torque will be explained using FIG. 1. FIG. 1 is an explanatory diagram illustrating spin torque.

As shown in FIG. 1, the magnetic tunnel junction structure has a reference layer, a barrier layer and a free layer. Also, electrons are flowing in various directions. Here, the spin torque is calculated using the magnetization vectors of the free layer and reference layer meshes that face each other with the barrier layer in between.

When modeling a magnetic tunnel junction structure by dividing it into meshes, the meshes of the free layer and the meshes of the reference layer must have a one-to-one correspondence. Here, one-to-one correspondence means that when the surface of the reference layer mesh in contact with the barrier layer is translated in parallel along the current vector (direction vector of the current flowing through the MTJ element), the surface of the reference layer mesh must match the surface of the free layer mesh that is in contact with the barrier layer.

FIG. 2 is an explanatory diagram illustrating mesh surfaces that have one-to-one correspondence. As shown in FIG. 2, the surface of the mesh of the reference layer in contact with the barrier layer is moved in parallel in the positive direction of the current vector, and thereby comes into contact with the surface of the mesh of the free layer in contact with the barrier layer. Note that the spin torque is calculated as the interaction of two meshes shown in FIG. 2.

Here, if the model is divided by triangular prism or hexahedral meshes, it is possible to easily make one-to-one correspondence between the free layer mesh and the reference layer mesh.

FIG. 3 is a diagram showing an example of a hexahedral mesh. As shown in FIG. 3, the free layer mesh 51, the barrier layer mesh 52, and the reference layer mesh 53 are hexahedral. In this case, the mesh 51 of the free layer and the mesh 53 of the reference layer have a one-to-one correspondence.

FIG. 4 is a diagram showing an example of a triangular prism mesh. As shown in FIG. 4, the free layer mesh 61, the barrier layer mesh 62, and the reference layer mesh 63 are triangular prisms. In this case, the mesh 61 of the free layer and the mesh 63 of the reference layer have a one-to-one correspondence.

On the other hand, if the model is divided by a tetrahedral mesh, it is not easy to make a one-to-one correspondence between the free layer mesh and the reference layer mesh. FIG. 5 is an explanatory diagram illustrating superposition of tetrahedral meshes. The surface 12S in FIG. 5 is the surface of the free layer in contact with the barrier layer when the model is divided into tetrahedral meshes. Furthermore, the surface 32S in FIG. 5 is the surface of the reference layer in contact with the barrier layer when the model is divided into tetrahedral meshes. Further, the triangles displayed on the surface 12S and the surface 32S are surfaces of a tetrahedral mesh.

As shown in FIG. 5, surfaces 12S and 32S cannot be matched exactly. In other words, when the surfaces 12S and 32S are superimposed, the surfaces of the tetrahedral meshes displayed on each do not match. In this way, when the model is divided by tetrahedral meshes, it is not easy to make one-to-one correspondence between the free layer mesh and the reference layer mesh.

[Functional configuration]
The configuration of the information processing device according to the embodiment will be explained using FIG. 6. FIG. 6 is a block diagram showing a configuration example of an information processing device. As shown in FIG. 6, the information processing device 10 includes an interface section 11, a storage section 12, and a control section 13. The information processing device 10 calculates the spin torque of each tetrahedral mesh in a model in which the magnetic storage element is divided into tetrahedral meshes.

The interface unit 11 is an interface for inputting and outputting data and communicating data with other devices. For example, the interface unit 11 may input and output data to and from input devices such as a keyboard and mouse, output devices such as a display and speakers, and external storage devices such as a USB memory. Further, for example, the interface unit 11 may be a NIC (Network Interface Card), and may perform data communication via the Internet.

The storage unit 12 is an example of a storage device that stores data, programs executed by the control unit 13, and the like, and is, for example, a hard disk or a memory. The storage unit 12 stores model information 121, corresponding mesh information 122, and magnetization vector information 123.

The model information 121 is information on a model in which the magnetic tunnel junction structure is divided into tetrahedral meshes. For example, the model information 121 is a mesh file. The model information 121 is information on a model as shown in FIG. FIG. 7 is a diagram showing an example of a model divided by a tetrahedral mesh. The model has a free layer 1L, a barrier layer 2L, and a reference layer 3L. Each layer is expressed as a set of tetrahedral meshes.

Furthermore, the model information 121 includes information specifying the triangular surfaces of the tetrahedral mesh on the surface 12S of the free layer 1L in contact with the barrier layer 2L and on the surface 32S of the reference layer 3L in contact with the barrier layer 2L. In the following explanation, the triangular face of the tetrahedral mesh will be referred to as a triangular mesh. The model information 121 includes the triangular mesh number and the coordinates of three nodes. Note that the surface 12S and the surface 32S can be said to be boundary surfaces between each layer.

FIG. 8 is a diagram showing an example of a triangular mesh. For example, the model information 121 includes information on triangular meshes on the surface 12S with numbers i-1, i, and i+1. Further, for example, the model information 121 includes information on triangular meshes on the surface 32S whose numbers are j-1, j, and j+1.

The corresponding mesh information 122 is data representing the correspondence between the virtual triangular prism mesh and the tetrahedral mesh for the free layer and the reference layer. FIG. 9 is a diagram showing an example of correspondence data between a virtual triangular prism mesh and a tetrahedral mesh on the free layer side. FIG. 10 is a diagram showing an example of correspondence data between a virtual triangular prism mesh and a tetrahedral mesh on the reference layer side. The virtual triangular prism mesh is a virtual triangular prism mesh created by a method described later. Note that the triangular prism mesh explained with reference to FIG. 4 is a mesh that actually constitutes a model, whereas the existence of a virtual triangular prism mesh is assumed for calculating spin torque and the like.

For example, FIG. 9 shows that among the tetrahedral meshes of the free layer 1L, those with numbers 2, 6, and 8 correspond to the virtual triangular prism mesh 1T(1). For example, FIG. 10 shows that among the tetrahedral meshes of the reference layer 3L, those with numbers 52, 57, 69, and 61 correspond to the virtual triangular prism mesh 1T(1).

The magnetization vector information 123 is the magnetization vector of each tetrahedral mesh. The magnetization vector information 123 is updated in spin torque calculation processing, which will be described later.

The control unit 13 executes a program stored in an internal storage device using RAM as a work area by, for example, a CPU (Central Processing Unit), MPU (Micro Processing Unit), GPU (Graphics Processing Unit), etc. This is realized by Further, the control unit 13 may be realized by, for example, an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array). The control unit 13 includes a reading unit 131, a creating unit 132, a specifying unit 133, a first calculating unit 134, and a second calculating unit 135.

The reading unit 131 reads data of a model in which each of the plurality of layers of the magnetic memory element is divided into tetrahedral meshes. For example, the reading unit 131 reads data of a model in which each of the free layer, barrier layer, and reference layer of the magnetic memory element is divided into tetrahedral meshes. The reading unit 131 stores the read model data in the storage unit 12 as model information 121. For example, the reading unit 131 associates the number identifying the triangular mesh of each layer with the position coordinates of each triangular mesh and stores them in the storage unit 12.

The creation unit 132 creates a virtual triangular prism mesh from the triangular mesh of the boundary surface of any layer when a plurality of layers are stacked, based on the data. For example, the creation unit 132 generates a virtual triangular prism mesh based on the locus of one of the triangular meshes that is part of the boundary surface and which creates a tetrahedral mesh in a direction perpendicular to the boundary surface. create.

In this embodiment, the creation unit 132 creates a first virtual triangular prism mesh from the interface between the free layer and the barrier layer when the free layer, the barrier layer, and the reference layer are stacked in this order, and Create a second triangular prism mesh from the boundary surface.

FIG. 11 is an explanatory diagram illustrating a method for creating a virtual triangular prism mesh. The triangular mesh 12t is a triangular mesh that exists on the surface 12S that is the interface between the free layer 1L and the barrier layer 2L. The creation unit 132 extrudes the triangular mesh 12t to the free layer 1L side by the same thickness as the free layer 1L. Next, the creation unit 132 extrudes the triangular mesh 12t to the reference layer 3L side by the same thickness as the barrier layer 2L, and then extrudes it by the same thickness as the reference layer 3L. Thereby, the creation unit 132 creates a virtual triangular prism mesh 1T. The virtual triangular prism mesh 1T is an example of a first virtual triangular prism.

Here, extruding means moving a surface in parallel. Therefore, the virtual triangular prism mesh 1T can be said to be a trajectory when the triangular mesh 12t is translated upward and downward. Further, the triangular mesh 12t' is a surface where the trajectory intersects with the surface 32S. Note that the upward direction is a direction from the triangular mesh 12t' to the triangular mesh 12t in FIG. 11. Further, it is assumed that the downward direction is the direction from the triangular mesh 12t to the triangular mesh 12t' in FIG.

Similarly, the creation unit 132 extrudes the triangular mesh on the surface 32S to the reference layer 3L side by the same thickness as the reference layer 3L. Next, the creation unit 132 extrudes the triangular mesh to the free layer 1L side by the same thickness as the barrier layer 2L, and then extrudes it by the same thickness as the free layer 1L. Thereby, the creation unit 132 creates a second triangular prism from the surface of the reference layer 3L. In this way, the creation unit 132 creates a virtual triangular prism mesh for each triangular mesh existing on the surface 12S and the surface 32S.

The identification unit 133 identifies tetrahedral meshes included in the virtual triangular prism mesh among the tetrahedral meshes that divide the plurality of layers. For example, the identifying unit 133 identifies a tetrahedral mesh whose center of gravity is included in the virtual triangular prism mesh, among the tetrahedral meshes that divide the plurality of layers. In this embodiment, the identifying unit 133 identifies the tetrahedral mesh included in each of the first virtual triangular prism mesh and the second virtual triangular prism mesh.

FIG. 12 is a diagram showing an example of a tetrahedral mesh included in a virtual triangular prism mesh. As shown in FIG. 12, the identifying unit 133 identifies the tetrahedral mesh 1p and the tetrahedral mesh 2p included in the free layer 1L as the tetrahedral mesh included in the virtual triangular prism mesh 1T. Furthermore, the specifying unit 133 also specifies the tetrahedral mesh included in the virtual triangular prism mesh 1T from the tetrahedral mesh included in the reference layer 3L. Similarly, the identifying unit 133 identifies the tetrahedral mesh included in the virtual triangular prism mesh created from the triangular mesh of the surface 32S.

The identifying unit 133 stores the number of the identified tetrahedral mesh in the storage unit 12 as the corresponding mesh information 122. Here, since a plurality of virtual triangular prism meshes 1T are created from the surface 12S, each virtual triangular prism mesh 1T is distinguished by a number.

The first calculation unit 134 uses the average of the magnetization vectors of the tetrahedral mesh specified by the identification unit 133 as the magnetization vector of the virtual triangular prism mesh, and calculates the spin torque of the virtual triangular prism mesh. In this embodiment, the first calculation unit 134 calculates the spin torque of each of the first virtual triangular prism mesh and the second virtual triangular prism mesh. The first calculation unit 134 can calculate the spin torque of the virtual triangular prism mesh using a known method.

As shown in FIGS. 9 and 10, the virtual triangular prism mesh 1T(1) includes the tetrahedral meshes of the free layer 1L with numbers 2, 6, and 8, and the reference meshes with numbers 52, 57, 69, and 61. A tetrahedral mesh of layer 3L is included. The first calculation unit 134 acquires the magnetization vector of each tetrahedron included in the virtual triangular prism mesh 1T(1) from the magnetization vector information 123, and calculates the average of the acquired magnetization vectors. Furthermore, the first calculation unit 134 regards the calculated average as the magnetization vector of the virtual triangular prism mesh 1T(1), and calculates the spin torque of the virtual triangular prism mesh 1T(1).

The second calculation unit 135 uses the spin torque of the virtual triangular prism mesh calculated by the first calculation unit 134 to calculate the spin torque of the tetrahedral mesh included in the virtual triangular prism mesh. For example, the second calculation unit 135 calculates the spin torque of the first virtual triangular prism mesh and the second virtual triangular prism mesh as the spin torque of the tetrahedral mesh included in both the first virtual triangular prism mesh and the second virtual triangular prism mesh. Calculate the average spin torque of .

As shown in FIG. 9, the tetrahedral mesh of the free layer 1L having the number 6 is included in both the virtual triangular prism mesh 1T(1) and the virtual triangular prism mesh 2T(1). Therefore, the second calculation unit 135 calculates the spin torque of the virtual triangular prism mesh 1T(1) and the spin torque of the virtual triangular prism mesh 2T(1) as the spin torque of the tetrahedral mesh of the free layer 1L with the number 6. Calculate the average.

[Processing flow]
FIG. 13 is a flowchart showing the flow of processing of the information processing device. As shown in FIG. 13, first, the information processing device 10 reads data of a model divided by a mesh (step S11). Here, if the read data is data of a model divided by a non-tetrahedral mesh (step S12, No), the information processing device 10 proceeds to step S14. On the other hand, if the read data is data of a model divided by a tetrahedral mesh (Step S12, Yes), the information processing device 10 performs a process of associating the virtual triangular prism mesh with the tetrahedral mesh (Step S13).

Furthermore, if the read data is data of a model divided by a non-tetrahedral mesh (step S14, No), the information processing device 10 calculates the spin torque of the non-tetrahedral mesh (step S16). For example, non-tetrahedrons are hexahedrons and triangular prisms.

On the other hand, if the read data is data of a model divided by a tetrahedral mesh (step S14, Yes), the information processing device 10 calculates the spin torque of the tetrahedral mesh (step S15).

Furthermore, the information processing device 10 calculates other magnetic fields (step S17), and updates the magnetization vector of each mesh stored in the magnetization vector information 123 (step S18). Here, the information processing device 10 calculates, for example, magnetic field interaction between meshes. Furthermore, if the number of repetitions is equal to or greater than the specified value (step S19, Yes), the information processing device 10 ends the process. On the other hand, if the number of repetitions is not equal to or greater than the specified value (step S19, No), the information processing device 10 increments the number of repetitions (step S20), and returns to step S14.

The details of the process of associating the virtual triangular prism mesh and the tetrahedral mesh (step S13 in FIG. 13) will be described with reference to FIG. 14. FIG. 14 is a flowchart showing the flow of processing for associating a virtual triangular prism mesh with a tetrahedral mesh.

As shown in FIG. 14, the information processing device 10 first identifies a triangular mesh at the interface between the free layer and the barrier layer (step S131). The information processing device 10 also identifies a triangular mesh at the interface between the reference layer and the barrier layer (step S132).

Next, the information processing device 10 creates a virtual triangular prism mesh by extrusion for each triangular mesh (step S133). Then, the information processing device 10 identifies the virtual triangular prism mesh included in each tetrahedral mesh, associates the tetrahedral mesh with the virtual triangular prism mesh, and stores them in the storage unit (step S134).

The details of the process of calculating the spin torque of the tetrahedral mesh (step S15 in FIG. 13) will be described using FIG. 15. FIG. 15 is a flowchart showing the flow of processing for calculating the spin torque of a tetrahedral mesh.

As shown in FIG. 15, the information processing device 10 first obtains the magnetization vector of the tetrahedral mesh (step S151). Next, the information processing device 10 calculates the magnetization vector of the virtual triangular prism mesh associated with the tetrahedral mesh based on the magnetization vector (step S152).

Here, the information processing device 10 loops steps S154 and S155 for each pair of virtual triangular prism meshes on which spin torque is applied (step S153). The information processing device 10 acquires the magnetization vector of the virtual triangular prism mesh included in the pair (step S154). Then, the information processing device 10 calculates the spin torque of the virtual triangular prism mesh included in the pair (step S155).

When the loop ends, the information processing device 10 calculates the spin torque of the tetrahedral mesh using the spin torque of the virtual triangular prism mesh (step S156).

[effect]
As described above, the reading unit 131 reads data of a model in which each of the plurality of layers of the magnetic memory element is divided into tetrahedral meshes. The creation unit 132 creates a triangular prism mesh from the triangular mesh of the boundary surface of any layer when a plurality of layers are stacked, based on the data. The identification unit 133 identifies tetrahedral meshes included in the triangular prism mesh among the tetrahedral meshes that divide the plurality of layers. The first calculation unit 134 uses the average of the magnetization vectors of the tetrahedral mesh specified by the identification unit 133 as the magnetization vector of the triangular prism mesh, and calculates the spin torque of the triangular prism mesh. The second calculation unit 135 uses the spin torque of the triangular prism mesh calculated by the first calculation unit 134 to calculate the spin torque of the tetrahedral mesh included in the triangular prism mesh. In this way, according to the information processing device 10 of this embodiment, the spin torque of the tetrahedral mesh can be obtained. As a result, it is possible to simulate the spin torque acting on the magnetic tunnel junction structure using a model in which the magnetic memory element is divided into tetrahedral meshes.

The creation unit 132 creates a triangular prism mesh as a locus when one of the triangular meshes that are part of the boundary surface and create a tetrahedral mesh is moved in a direction perpendicular to the boundary surface. Thereby, the information processing device 10 can create a triangular prism mesh using the faces of the tetrahedral mesh.

The identifying unit 133 identifies a tetrahedral mesh whose center of gravity is included in a triangular prism mesh, among the tetrahedral meshes that divide a plurality of layers. Thereby, the information processing device 10 can easily associate the tetrahedral mesh with the triangular prism mesh, regardless of the size or shape of the tetrahedral mesh.

The reading unit 131 reads data of a model in which each of the free layer, barrier layer, and reference layer of the magnetic memory element is divided into tetrahedral meshes. The creation unit 132 creates a first triangular prism mesh from the interface between the free layer and the barrier layer when the free layer, barrier layer, and reference layer are stacked in this order, and creates a second triangular prism mesh from the interface between the barrier layer and the reference layer. Create a triangular prism mesh. The specifying unit 133 specifies the tetrahedral mesh included in each of the first triangular prism mesh and the second triangular prism mesh. The first calculation unit 134 calculates the spin torque of each of the first triangular prism mesh and the second triangular prism mesh. The second calculation unit 135 calculates the spin torque of the first triangular prism mesh and the spin torque of the second triangular prism mesh as the spin torque of the tetrahedral mesh included in both the first triangular prism mesh and the second triangular prism mesh. Calculate the average. In this way, according to the information processing device 10 of this embodiment, the spin torque of the tetrahedral mesh that divides the magnetic storage element consisting of the free layer, barrier layer, and reference layer can be obtained. As a result, it is possible to simulate the spin torque acting on the magnetic tunnel junction structure using a model in which a magnetic memory element consisting of a free layer, a barrier layer, and a reference layer is divided into tetrahedral meshes.

[system]
Information including processing procedures, control procedures, specific names, and various data and parameters shown in the above documents and drawings can be changed arbitrarily unless otherwise specified. Furthermore, the specific examples, distributions, numerical values, etc. described in the examples are merely examples, and can be changed arbitrarily.

Furthermore, each component of each device shown in the drawings is functionally conceptual, and does not necessarily need to be physically configured as shown in the drawings. That is, the specific form of distributing and integrating each device is not limited to what is shown in the drawings. In other words, all or part of them can be functionally or physically distributed and integrated into arbitrary units depending on various loads, usage conditions, and the like. Furthermore, all or any part of each processing function performed by each device can be realized by a CPU and a program that is analyzed and executed by the CPU, or can be realized as hardware using wired logic.

[hardware]
FIG. 16 is a diagram illustrating an example of a hardware configuration. As shown in FIG. 16, the information processing device 10 includes a communication interface 10a, an HDD (Hard Disk Drive) 10b, a memory 10c, and a processor 10d. Furthermore, the parts shown in FIG. 16 are interconnected by a bus or the like.

The communication interface 10a is a network interface card or the like, and communicates with other servers. The HDD 10b stores programs and DB that operate the functions shown in FIG.

The processor 10d reads a program that executes the same processing as each processing unit shown in FIG. It is a hardware circuit. That is, this process executes the same functions as each processing unit included in the information processing device 10. Specifically, the processor 10d reads a program having the same functions as the reading section 131, the creating section 132, the specifying section 133, the first calculating section 134, and the second calculating section 135 from the HDD 10b or the like. The processor 10d then executes a process that performs the same processing as the reading unit 131, the creating unit 132, the identifying unit 133, the first calculating unit 134, the second calculating unit 135, and the like.

In this way, the information processing device 10 operates as an information processing device that executes a learning method by reading and executing a program. Further, the information processing device 10 can also realize the same functions as in the above-described embodiments by reading the program from the recording medium using the medium reading device and executing the read program. Note that the programs in other embodiments are not limited to being executed by the information processing device 10. For example, the present invention can be similarly applied to cases where another computer or server executes a program, or where these computers or servers cooperate to execute a program.

This program can be distributed via a network such as the Internet. Additionally, this program is recorded on a computer-readable recording medium such as a hard disk, flexible disk (FD), CD-ROM, MO (Magneto-Optical disk), or DVD (Digital Versatile Disc), and is read from the recording medium by the computer. It can be executed by being read.

1L Free layer 2L Barrier layer 3L Reference layer 12S, 32S Surface 1T, 2T Virtual triangular prism mesh 1p, 2p Tetrahedral mesh 12t Triangular mesh 10 Information processing device 11 Interface section 12 Storage section 13 Control section 121 Model information 122 Corresponding mesh information 123 Magnetization Vector information 131 Reading section 132 Creation section 133 Specification section 134 First calculation section 135 Second calculation section

Claims (5)

a reading unit that reads data of a model in which each of the plurality of layers of the magnetic memory element is divided into tetrahedral meshes;
a creation unit that creates a triangular prism mesh from a triangular mesh of a boundary surface of one of the layers when the plurality of layers are stacked, based on the data;
a specifying unit that identifies a tetrahedral mesh whose center of gravity is included in the triangular prism mesh among the tetrahedral meshes that divide the plurality of layers;
a first calculation unit that calculates the spin torque of the triangular prism mesh by setting the average of the magnetization vectors of the tetrahedral mesh specified by the identification unit as the magnetization vector of the triangular prism mesh;
a second calculation unit that calculates a spin torque of a tetrahedral mesh included in the triangular prism mesh using the spin torque of the triangular prism mesh calculated by the first calculation unit;
An information processing device comprising: The creation unit generates a locus of one of the triangular meshes, which is a part of the boundary surface and creates the tetrahedral mesh, in a direction perpendicular to the boundary surface, into the triangular prism mesh. The information processing apparatus according to claim 1, wherein the information processing apparatus is created as: The reading unit reads data of a model in which each of the free layer, barrier layer, and reference layer of the magnetic memory element is divided into tetrahedral meshes,
The creation unit creates a first triangular prism mesh from the interface between the free layer and the barrier layer when the free layer, the barrier layer, and the reference layer are stacked in this order, and Create a second triangular prism mesh from the layer interface,
The identifying unit identifies a tetrahedral mesh included in each of the first triangular prism mesh and the second triangular prism mesh,
The first calculation unit calculates spin torque of each of the first triangular prism mesh and the second triangular prism mesh,
The second calculation unit calculates the spin torque of the first triangular prism mesh and the second triangular prism mesh as the spin torque of the tetrahedral mesh included in both the first triangular prism mesh and the second triangular prism mesh. 2. The information processing apparatus according to claim 1, wherein the information processing apparatus calculates an average of spin torques. a reading unit that reads data of a model in which each of the plurality of layers of the magnetic memory element is divided into tetrahedral meshes;
a creation unit that creates a triangular prism mesh from a triangular mesh of a boundary surface of one of the layers when the plurality of layers are stacked, based on the data;
a specifying unit that identifies a tetrahedral mesh whose center of gravity is included in the triangular prism mesh among the tetrahedral meshes that divide the plurality of layers;
a first calculation unit that calculates the spin torque of the triangular prism mesh by setting the average of the magnetization vectors of the tetrahedral mesh specified by the identification unit as the magnetization vector of the triangular prism mesh;
a second calculation unit that calculates a spin torque of a tetrahedral mesh included in the triangular prism mesh using the spin torque of the triangular prism mesh calculated by the first calculation unit;
A magnetic memory element simulator system comprising: Loads data from a model in which each of the multiple layers of the magnetic memory element is divided into tetrahedral meshes,
Based on the data, create a triangular prism mesh from the triangular mesh of the boundary surface of any layer when the plurality of layers are stacked,
Among the tetrahedral meshes that divide the plurality of layers, identify a tetrahedral mesh whose center of gravity is included in the triangular prism mesh,
The average of the magnetization vectors of the identified tetrahedral mesh is set as the magnetization vector of the triangular prism mesh, and the spin torque of the triangular prism mesh is calculated,
A magnetic storage element simulator method, characterized in that a computer executes a process of calculating a spin torque of a tetrahedral mesh included in the triangular prism mesh using the calculated spin torque of the triangular prism mesh.

Images