JPH11329932A - Light intensity simulation apparatus, light intensity simulation method, and recording medium recording light intensity simulation program - Google Patents

Abstract

(57) [Summary] [PROBLEMS] A light intensity simulation apparatus capable of efficiently simulating an enormous amount of mask patterns and performing optimized parallel processing by utilizing a plurality of arithmetic processors not specially designed. A method and a recording medium on which a light intensity simulation program is recorded. SOLUTION: When pattern data and optical conditions are inputted from an input device 8, a control processor 1
The required calculation amount in the light intensity simulation based on the various data input is obtained, and the calculation amount of the light intensity simulation is allocated to each of the arithmetic processors in the arithmetic processor group 7. In the arithmetic processing processor group 7,
Each arithmetic processing processor performs parallel processing on the allocated calculation amount, and the control processor 1 sums and averages the result and outputs the result to the monitor 9, the printer 10, and the like.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a light intensity simulation for performing a simulation of an exposure process, which is mainly used in research and development of a photomask pattern and as a verification means in a design process in a photolithography process in a semiconductor manufacturing process. The present invention relates to an apparatus, a light intensity simulation method, and a recording medium on which a light intensity simulation program is recorded.

[0002]

2. Description of the Related Art Conventionally, in the research and development or development prototype stage of a semiconductor process, computer simulation has been used as a technique for grasping the characteristics of the process and the product, and for virtually testing and estimating the characteristics with respect to the manufacturing conditions. There are technologies, and they are currently used as indispensable technologies for semiconductor design. In particular, the simulation technology of the photolithography process, which is the main microfabrication technology in semiconductor manufacturing technology, has been theoretically established and is an indispensable technology in research and development. Among the simulations of photolithography, the simulation of the exposure process is particularly called “light intensity simulation”, and a photomask pattern (hereinafter, referred to as a mask pattern) is exposed and transferred onto a wafer using a projection exposure apparatus (also referred to as a stepper). The light intensity distribution of the projected optical image at this time is obtained by calculation.

[0003] The theory underlying light intensity simulation technology is a physical theory based on imaging optics theory established by H. Hopkins et al. (Reference: Born and Wolf, "Principles of Optics II and III" 1975, and H.Hopkins; J.Opt.Soc.Am, V
ol. 47, No. 6 ('57) p508-, etc.), and a computer calculation model includes a model by Lin or Yeung. Software for performing computer simulation is also called a simulator.

Since the light intensity simulation can estimate the exposure distribution on the wafer without actually performing lithography, the light intensity simulation has been frequently used in the research and development of the lithography process and in the trial manufacture of devices. In particular, in recent years, the required fine processing technology is approaching the limit of processing by light, and it has become difficult to develop devices by actually performing experiments, both technically and costly. Simulation techniques that can be used to obtain results quickly and at low cost have become increasingly important.

[0005]

At present, attention has been paid to so-called ultrafine processing techniques such as a super-resolution technique such as a phase shift method and an optical proximity correction (OPC) technique as advanced techniques of lithography. However, all of these require processing of a mask pattern in accordance with a rule for optimizing a lithography process different from a design rule of an LSI circuit. For this reason, in order to optimize the mask pattern, it is necessary to use a lithography process, or at least an optimization unit that takes into account the exposure process. For that purpose, the light intensity simulation is used to optimize the pattern based on the exposure conditions. Means were needed.

The light intensity simulation is performed in a semiconductor LSI.
The optical system of the reduction projection exposure apparatus (stepper) generally used in the manufacturing process is configured as a physical model,
The configuration of the optical system mainly includes a light source, an illumination lens, a photomask, a projection lens, and a wafer. Although details are given in the above-mentioned literature, the light intensity simulation simulates these optical systems numerically, and the mask pattern image projected and exposed is transferred onto the wafer by calculation using the above-mentioned imaging optical theory. This is to obtain a projected image at the time of occurrence.

Further, since the light intensity simulation uses an optical theory, there are less uncertainties and a highly accurate result can be obtained as compared with the simulation of other lithography processes which need to calculate the behavior at the molecular and atomic levels. Therefore, attempts to feed back simulation results to pattern design have become active in recent years. However, light intensity simulation generally requires a huge amount of calculation, so even with the latest high-speed computer,
It is difficult to simulate the entire pattern of the LSI chip. In particular, in the light intensity simulation, since the calculation amount increases in proportion to the square to the fourth power of the pattern area (the calculation amount differs depending on the calculation algorithm), it is practically possible to calculate only a pattern with a small area. There was a defect.

On the other hand, as a method of performing a simulation with a large amount of calculation, attempts have been made to connect a plurality of arithmetic processing processors and perform parallel processing.
In many cases, unless the device is optimized and designed as a dedicated parallel processing device, efficient operation processing cannot be performed in terms of algorithm, so that the device becomes expensive and the operation efficiency is poor.

Further, the actual LSI pattern data has a very large area and is complicated, and is usually composed of hundreds of thousands to tens of millions of huge closed figures. There is also a prospect of further expansion. For a pattern having such a huge data amount, it is practically impossible to perform light intensity simulation on the entire mask pattern in order to optimize the fine processing accuracy from the viewpoint of time and cost. There is also a method in which light intensity simulation is performed only on a limited partial area. However, in this method, an appropriate pattern must be extracted manually, and practical use in the mass production stage is extremely difficult. .

In view of the above circumstances, the present invention efficiently simulates an enormous amount of mask patterns and can perform optimized parallel processing by utilizing a plurality of arithmetic processors which are not specially designed. An object of the present invention is to provide a light intensity simulation device, a light intensity simulation method, and a recording medium on which a light intensity simulation program is recorded.

[0011]

According to a first aspect of the present invention, there is provided a photomask pattern based on predetermined data based on pattern data of the photomask pattern and optical condition parameters obtained by parameterizing various optical conditions in an exposure process of the photomask pattern. In a light intensity simulation apparatus that performs an arithmetic process according to a calculation algorithm and obtains a light intensity distribution of a projection optical image when the photomask pattern is exposed and transferred onto a wafer, the pattern data and a numerical range of the optical condition parameter And data input means for inputting the variable step amount, and calculating the amount of calculation required for the light intensity simulation based on the input pattern data and the numerical range of the optical condition parameter and the variable step amount, and distributing the calculation amount to a plurality. And a plurality of arithmetic processors The light intensity simulation by synthesizing parallel processing means for performing parallel processing in each processing processor on each calculation amount distributed by the calculation amount distribution means in each processing processor; A light intensity simulation apparatus comprising: a simulation result calculation means for calculating a result.

According to a second aspect of the present invention, in the light intensity simulation apparatus according to the first aspect, the predetermined calculation algorithm is a numerical integration calculation based on the pattern data and the value of the optical system parameter. hand,
The numerical value integration calculation results when the optical system parameters are changed by the variable step amount within the numerical value range are summed and averaged, and the calculation amount distribution means includes a numerical value of the input optical condition parameters. From the range and the variable step amount, the number of combinations of the values of the optical condition parameters used in the numerical integration calculation is obtained, and the obtained number of combinations of the optical system parameters is calculated by a plurality of arithmetic processors And each of the arithmetic processing processors of the parallel processing means performs the numerical integration based on the combination of the pattern data and the values of various optical system parameters distributed by the calculation amount distribution means. It is characterized by performing calculations.

According to a third aspect of the present invention, in the light intensity simulation apparatus according to the second aspect, the calculation amount distributing means sets the distribution amount of the number of combinations of the values of the various optical system parameters to the plurality of the plurality of optical system parameters. It is characterized in that it is changed according to each processing performance of the processing processor.

According to a fourth aspect of the present invention, there is provided an arithmetic operation according to a predetermined calculation algorithm based on pattern data of a photomask pattern and optical condition parameters obtained by parameterizing various optical conditions in an exposure process of the photomask pattern. In a light intensity simulation apparatus for performing processing and obtaining a light intensity distribution of a projection optical image when the photomask pattern is exposed and transferred onto a wafer, the light intensity simulation apparatus includes data communication means connected to a network and capable of exchanging data with each other. A plurality of computers, one of the plurality of computers, the data input means for inputting the pattern data, the numerical range of the optical condition parameter and the variable step amount,
Calculating a calculation amount required for the light intensity simulation based on the numerical range and the variable step amount of the input pattern data and the optical condition parameter, and having a calculation amount distribution unit for distributing the calculation amount to a plurality of; Each of the other computers, except for one computer, performs the arithmetic processing of each calculation amount distributed by the calculation amount distribution means in each arithmetic processing processor in parallel.
A light intensity simulation apparatus characterized in that one computer combines the results of the arithmetic processing performed by the other computers except for the above, and obtains the light intensity simulation result.

According to a fifth aspect of the present invention, an arithmetic operation according to a predetermined calculation algorithm is performed based on pattern data of a photomask pattern and optical condition parameters obtained by parameterizing various optical conditions in an exposure process of the photomask pattern. Performing a process, a light intensity simulation method for obtaining a light intensity distribution of a projection optical image when the photomask pattern is exposed and transferred onto a wafer, wherein the pattern data, the numerical range of the optical condition parameters and the variable step amount A first step of inputting, and a second step of obtaining a calculation amount required for light intensity simulation based on the numerical range and the variable step amount of the input pattern data and optical condition parameters, and distributing the calculation amount to a plurality of steps. , Each of the plurality of arithmetic processors is assigned to the plurality of processors. A third step of parallel processing operations according to the calculation algorithm by computational minutes, separated,
A fourth step of combining the results of the arithmetic processing performed in parallel by the plurality of arithmetic processors and obtaining the light intensity simulation result.

According to a sixth aspect of the present invention, in the light intensity simulation method according to the fifth aspect, the predetermined calculation algorithm is a numerical integration calculation based on the pattern data and the value of the optical system parameter. hand,
Sums and averages respective numerical integration calculation results when the optical system parameter is changed by the variable step amount within the numerical value range. In the second step, numerical values of the input optical condition parameters are set. From the range and the variable step amount, the number of combinations of the values of the optical condition parameters used in the numerical integration calculation is obtained, and the obtained number of combinations of the optical system parameters is reduced to the number of the plurality of arithmetic processors. In the third stage, each of the plurality of arithmetic processors is based on the combination of the pattern data and the values of various optical system parameters distributed in the second stage.
It is characterized in that the numerical integration calculation is performed.

According to a seventh aspect of the present invention, in the light intensity simulation method according to the sixth aspect, in the second step, the parameters of the various optical system parameters distributed according to the number of the plurality of arithmetic processing processors are provided. It is characterized in that the number of value combinations is changed in accordance with each processing performance of the plurality of processing processors.

The invention according to claim 8 is an arithmetic operation according to a predetermined calculation algorithm based on pattern data of a photomask pattern and optical condition parameters obtained by parameterizing various optical conditions in an exposure process of the photomask pattern. A computer-readable recording medium having recorded thereon a light intensity simulation program for obtaining a light intensity distribution of a projection optical image when the photomask pattern is exposed and transferred onto a wafer, the light intensity simulation program comprising: A first procedure for inputting the pattern data, the numerical range and the variable step amount of the optical condition parameter to the computer, and a first procedure for inputting the pattern data and the numerical range and the variable step amount of the optical condition parameter. Light intensity simulation A second procedure for determining the amount of calculation required for the computer and distributing the amount of calculation to a plurality of processors, and instructing each of the plurality of arithmetic processing processors of the computer by the amount of calculation allocated to the plurality of processors in accordance with the calculation algorithm. And a fourth step of combining the respective arithmetic processing results performed in parallel by the plurality of arithmetic processors to obtain the light intensity simulation result. It is a computer-readable recording medium on which an intensity simulation program is recorded.

According to a ninth aspect of the present invention, in the computer readable recording medium having recorded thereon the light intensity simulation program according to the eighth aspect, the predetermined calculation algorithm in the light intensity simulation program includes the pattern data, And a numerical integration calculation based on the value of the optical system parameter, wherein the numerical integration calculation results when the optical system parameter is changed by the variable step amount within the numerical range are summed and averaged, In the second procedure, the number of combinations of the values of the optical condition parameters used for the numerical integration calculation is calculated from the numerical range and the variable step amount of the input optical condition parameters. The number of combinations of optical system parameters Partitioned according to, in the third procedure, each of the plurality of arithmetic processors, the pattern data, and, based on a combination of the values of the various optical systems parameters was partitioned second procedure,
It is characterized in that the numerical integration calculation is performed.

According to a tenth aspect of the present invention, there is provided a computer-readable recording medium having recorded thereon the light intensity simulation program according to the ninth aspect, wherein in the second step, the number of the plurality of arithmetic processors is changed. The number of combinations of the values of the various optical system parameters to be distributed is changed according to each processing performance of the plurality of processing processors.

Here, in each of the above-mentioned claims, the pattern data means, for example, a photomask pattern to be subjected to light intensity simulation is a grid of a predetermined size.
Mesh), and the size of this mesh,
It consists of the numerical value of the light transmittance (complex transmittance to represent the intensity transmittance and the phase of light) for each mesh and the dimensions of the photomask. The various optical system parameters include, for example, a light source wavelength, a lens numerical aperture, and a coherence factor (=
Coherency), an out-of-focus value (= defocus), and the like.

In the third, seventh, and tenth aspects, changing the distribution amount of the number of combinations of the values of various optical system parameters in accordance with the respective processing performances of the plurality of processing processors means, for example, Assuming that there are ten arithmetic processors having the same arithmetic processing capability and that the number of combinations of various optical system parameter values is 30, there are three sets of various optical system parameter values distributed to each arithmetic processor. (Ie, evenly distributed), but assume that there are five arithmetic processors with a certain arithmetic processing capability and five arithmetic processors with an arithmetic processing capability twice that of this arithmetic processor (that is, 10 arithmetic processing processors), and assuming that there are 30 combinations of values of various optical system parameters, it has twice the arithmetic processing capability. The five arithmetic processor that,
This means that four sets of values of various optical system parameters are distributed, and two sets of values are distributed to the remaining five arithmetic processing processors.

That is, the distribution amount of the number of combinations is adjusted according to the arithmetic performance of each processor so that the time required for arithmetic processing in one processor is the same for all processors. This is because if the number of combinations and the computational performance of the processor are known in advance, the distribution amount can be obtained by a simple calculation.

With this configuration, when pattern data is input, the calculation range of the light intensity simulation determined from the size of the pattern data and the optical conditions is determined, and the calculation range is determined according to the number of available processors. After determining the distribution of the calculation range, the respective calculation processors calculate the respective calculation ranges in parallel based on the determination, and then combine the calculation results to obtain a desired simulation result.

Further, since the simulation calculation by the parallel processing is performed, the simulation with the optimized efficiency can always be performed, and the processing time can be greatly reduced. Further, since the parallel processing at this time simply allocates the calculation range, any computer connected by a network can be used.

[0026]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the contents of the present invention will be described in detail with reference to the accompanying drawings. FIG. 1 is a diagram showing a configuration of a light intensity simulation apparatus according to the present invention. In the figure, a control processor 1 controls each unit of a light intensity simulation device (hereinafter, abbreviated as “this device”) in the present embodiment via a bus B. Reference numeral 2 denotes a ROM, which stores a startup program, a basic operation program, and the like of the apparatus. Reference numeral 3 denotes a RAM which stores a light intensity simulation program according to the present embodiment installed in the apparatus. Further, data generated during the process of executing the light intensity simulation is temporarily stored.

4 is a hard disk drive (HD)
D), and stores data such as the light intensity simulation program and the result of the light intensity simulation in the present embodiment. 5 is a floppy disk, C
This is a computer-readable recording medium such as a D-ROM, an MO disk, and a semiconductor memory, and stores the light intensity simulation program according to the present embodiment.

Reference numeral 6 denotes a data reading / writing device for reading a program or input data recorded on the recording medium 5 and data such as a light intensity simulation result stored in the RAM 3 and the HDD 4 to the recording medium 5. Is written. 7 is a group of arithmetic processing processors,
.. Cn connected to each other by a local bus LB, and performs arithmetic processing at the time of light intensity simulation under the control of the control processor 1.

Reference numeral 8 denotes an input device including a mouse, a keyboard, and the like, which performs operations and the like of the light intensity simulation device according to the present embodiment. A monitor 9 displays an operation screen, a data input screen, a light intensity simulation result screen, and the like of the light intensity simulation apparatus according to the present embodiment. A printer 10 prints out a light intensity simulation result based on an instruction from an operator.

Next, the operation of the light intensity simulation apparatus according to this embodiment will be described with reference to the flowchart shown in FIG. First, the operator sets the recording medium 5 on which the program for executing the light intensity simulation according to the present embodiment is recorded in the data reading / writing device 4 of FIG. 1, and installs the program on the device (for example, The program recorded on the recording medium 5 is copied to a predetermined storage area of the HDD 4 or the like).

When the operator starts the installed program, the program is stored in the HDD.
4 is read into the RAM 2, and thereafter, the control processor 1 starts the processing of the flowchart shown in FIG. 2 according to the program read into the RAM 2. First, the data input process of step S1 is performed. That is,
A data input screen for inputting input data used for performing the light intensity simulation is displayed on the monitor 9 of the apparatus. Thus, the operator uses the input device 8 to input data used in the light intensity simulation on the data input screen displayed on the monitor 9. Therefore, the input device 8 can be said to be data input means.

Here, as input data, data relating to a mask pattern (hereinafter referred to as pattern data)
And various optical system parameter data (hereinafter, referred to as optical system parameters) of the exposure apparatus assumed in performing the light intensity simulation. The pattern data is, for example, when there is a mask pattern P (M is a mask portion in the figure) as shown in FIG.
Is divided into a grid of a predetermined size (hereinafter, referred to as a mesh), and the light transmittance of each mesh m (exactly,
The value of the intensity transmittance and the complex transmittance is set to represent the phase of light. Further, the pattern area, that is, the dimension X,
The values of Y and the mesh widths x and y are also input as pattern data.

The optical system parameters are numerical values representing parameters of an optical system in an assumed exposure apparatus, and generally include a light source wavelength, a lens numerical aperture, a coherence (= coherency), and a defocus value. (= Defocus)
Are used, and the optical system parameters input at the time of the light intensity simulation are a variable range and a variable step of these parameters.

Next, proceeding to step S2, the control processor 1 performs pattern data processing based on the input data in order to perform an efficient operation by parallel processing enabled by using a calculation algorithm of light intensity simulation. From the values of the size, that is, the area of the pattern, and the value of the mesh width, and the optical system parameters, that is, the light source wavelength, the lens numerical aperture, the coherence, and the defocus value,
The calculation range of the light intensity simulation is obtained.

Here, the calculation range of the light intensity simulation will be described. The calculation algorithm of the light intensity simulation is basically called Fourier integration,
It consists of a repetition of some kind of numerical integration (for details, see the above-mentioned reference: Born, Wolf, Principles of Optics II ・ I
II, 1975, and H. Hopkins; J. Opt. Soc. Am, Vol. 47,
No. 6 ('57) p508-). That is, the result of the light intensity simulation is obtained by repeating the numerical integration calculation while changing the optical system parameters little by little (or by the change set by the operator) based on the fixed pattern data, and taking the sum average thereof. Results are required.

The numerical integrations calculated by substituting the respective optical system parameters are independent of each other, so that the calculation in each arithmetic processing processor can be executed without interference. That is, based on fixed pattern data, that is, the light transmittance of each mesh, numerical integration calculations in which the values of the optical system parameters differ little by little (or by the amount of change set by the operator) are individually executed and individually obtained. The result is obtained by taking the sum average of the numerical integration results.

Therefore, the calculation range referred to here is the total numerical integration calculation amount processed for performing the light intensity simulation based on the input data.
This calculation amount can be determined from the size of the pattern data and the number of combinations of the optical system parameters.

Next, based on the calculation range obtained in step S2, the control processor 1 gives each of the arithmetic processors c1, c2,... Make a distribution. That is, first, the control processor 1 sends each of the arithmetic processors c1 and c via the local bus LB.
, Cn, the pattern data input in step S1 (common to each processor) is sent out, and then each processor c1, c
According to the arithmetic processing capacity of 2,..., Cn, a set of parameters to be substituted into the above-described numerical integration is divided into an appropriate number and transmitted. Thus, the control processor 1 can be referred to as a calculation amount distribution unit.

At this time, if the arithmetic processors in the arithmetic processor group 7 have the same arithmetic performance, the control processor 1 sets a set of parameters to be sent to each of them so that the processing load of numerical integration becomes equal. Make the numbers even. For example, n arithmetic processing processors c1, c
When the computational performance of 2,..., Cn is equivalent and the number of parameter sets is p, the number of numerical integration processes per computation processor is p / n (when p is divisible by n)
It is. Therefore, the processing amount is reduced to 1 / n, that is, the processing time is reduced to 1 / n as compared with the case where the calculation is performed by only one arithmetic processing processor.

If p / n is not divisible, the processing processor may or may not process surplus data. However, the difference in the number of processed data is at most one. There is no significant drop.

When there is a difference in the arithmetic performance between the arithmetic processors, for example, the number of data to be processed (that is, the number of sets of parameters to be sent to each arithmetic processor) is adjusted according to the performance. Thus, the processing time of the entire arithmetic processing processor may be adjusted to be as equal as possible.

Thus, each of the processors in the group of processors 7 capable of performing parallel processing, based on the pattern data sent from the control processor 1 and the set of optical system parameters, respectively, The numerical integration calculation is executed independently (step S4). Therefore, it can be said that the processor group 7 is a parallel processing means.

Next, proceeding to step S5, the control processor 1 collects calculation result data calculated in each of the arithmetic processors of the arithmetic processor group 7,
Take the sum average of them. Here, since the calculation result calculated by each arithmetic processor is numerical matrix data having the same number of arrays as the pattern data, the sum total average simply takes the sum of the elements of the matrix, and calculates the sum of the optical system parameters. It can be obtained by dividing by a number (ie, the number of repetitions of numerical integration) and averaging. This allows
Light intensity distribution data is obtained as a light intensity simulation result. Thereby, the control processor 1
Can be said to be a simulation result calculating means as well as a calculation amount distributing means as described above.

In step S6, the control processor 1 performs output processing of the light intensity simulation result. That is, the result of the light intensity simulation is displayed on the screen of the monitor 9. When the operator instructs to print out the result of the light intensity simulation via the input device 8, print data of the light intensity simulation result is output to the printer 10 and the result of the light intensity simulation is output. Print out.

Further, according to an instruction from the operator, the result of the light intensity simulation is stored in the HDD 4 or the recording medium 5 for recording data (a recording medium different from the recording medium in which the light intensity simulation program of the present embodiment is recorded). And save it as a data file.

As described above, according to the light intensity simulation apparatus of the present embodiment, the numerical integration calculation for performing the light intensity simulation is equally distributed to each of the arithmetic processing processors capable of parallel processing. Efficient parallel processing can be performed, and processing time is reduced.

As described above, the data input processing in step S1 of FIG. 2 includes, in addition to the method of inputting from the input device 8 of FIG. 5
(However, the program may be stored in a recording medium different from the recording medium on which the light intensity simulation program of the present embodiment is recorded), and the data file specified by the operator may be read from these storage devices or the recording medium. good. Therefore, in this case, the HDD 4 or the storage medium 5 and the control processor 1 can be regarded as data input means.

In the above-described light intensity simulation apparatus, the light intensity simulation program recorded on the recording medium 5 is once read into the HDD 4 or the RAM 3 and then processed by the control processor 1. Simulation program in ROM
2 may be stored. That is, in this case, the ROM 2 can be regarded as a computer-readable recording medium that records the light intensity simulation program according to the present embodiment.

Another embodiment may be a so-called multi-processor machine designed for parallel processing. In this case, each processor receives data at the same time and executes calculations at the same time. Is possible. Instead of a single device, a plurality of computers in a general network computer distributed processing environment may be used.

That is, as shown in FIG. 4, for example, a plurality of computers 21-1 to 21-n connected to the network 20 and capable of performing data communication with each other are used, and one of the computers (for example, the computer 21-1) is used. )
Are the processes other than the above-described arithmetic processing processor group 7 in FIG. 1 (in FIG. 2, steps S1 to S3 and step S1).
5, S6).

In addition, software carrying the same light intensity simulation algorithm is installed in other computers (for example, computers 21-2 to 21-n), and each computer transmits the same algorithm from the computer 21-1. Numerical integration calculation (the process of step S4 in FIG. 2) may be executed based on a set of the obtained pattern data and light intensity parameters. With such a configuration, it can be executed regardless of the type of computer or operation system.

[0052]

As described above, according to the apparatus configuration as in the present invention, if pattern data is input, the optimal distribution of the light intensity simulation calculation range is determined according to the number of arithmetic processing processors and the arithmetic performance. By executing the parallel processing based on the calculation, efficient simulation calculation can be performed, and the calculation time can be greatly reduced. In addition, since this device configuration can easily convert a computer connected by a network, a very flexible and appropriate configuration can be taken according to the usage environment of the computer.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a light intensity simulation apparatus according to an embodiment of the present invention.

FIG. 2 is a flowchart showing a procedure of processing in the light intensity simulation apparatus.

FIG. 3 is an explanatory diagram for explaining the contents of pattern data used in calculation processing in the light intensity simulation apparatus.

FIG. 4 is a schematic configuration diagram showing a schematic configuration of another embodiment of the light intensity simulation apparatus according to the present invention.

[Description of sign]

 DESCRIPTION OF SYMBOLS 1 Control processor 2 ROM 3 RAM 4 HDD (hard disk drive) 5 Recording medium 6 Data read / write device 7 Arithmetic processing processor group 8 Input device 9 Monitor 10 Printer 20 Network 21-1 to 21-n Computer

Claims (10)

[Claims] An arithmetic process according to a predetermined calculation algorithm is performed based on pattern data of a photomask pattern and optical condition parameters obtained by parameterizing various optical conditions in an exposure process of the photomask pattern,
In a light intensity simulation apparatus for obtaining a light intensity distribution of a projection optical image when the photomask pattern is exposed and transferred onto a wafer, a data input for inputting the pattern data, a numerical range of the optical condition parameter, and a variable step amount Means for calculating a calculation amount required for light intensity simulation based on a numerical range and a variable step amount of the input pattern data and optical condition parameters, and distributing the calculation amount to a plurality of calculation amount distributing means; A parallel processing unit having a processing processor and performing parallel processing in each processing processor of each calculation amount distributed by the calculation amount distribution unit; and synthesizing each calculation processing result performed by the parallel processing unit. Simulation result calculation for calculating the light intensity simulation result And a light intensity simulation device. 2. The predetermined calculation algorithm is a numerical integration calculation based on the pattern data and the value of the optical system parameter, wherein the predetermined calculation algorithm changes the optical system parameter by the variable step amount within the numerical value range. The calculation amount distributing means, based on the numerical value range and variable step amount of each of the input optical condition parameters, calculates each of the respective numerical integration calculation results. Calculating the number of combinations of the values of the optical condition parameters, distributing the calculated number of combinations of the optical system parameters according to the number of the plurality of arithmetic processors in the parallel processing means, Is a set of the pattern data and the value of various optical system parameters distributed by the calculation amount distribution means. Light intensity simulation device according to claim 1, characterized in that the numerical integral calculation on the basis of the total. 3. The calculation amount distribution unit changes a distribution amount of the number of combinations of the values of the various optical system parameters in accordance with each processing performance of the plurality of processing processors. Item 3. A light intensity simulation apparatus according to item 2. 4. An arithmetic processing according to a predetermined calculation algorithm is performed based on pattern data of a photomask pattern and optical condition parameters obtained by parameterizing various optical conditions in an exposure process of the photomask pattern.
In a light intensity simulation device for obtaining a light intensity distribution of a projection optical image when the photomask pattern is exposed and transferred onto a wafer, a plurality of computers each having a data communication unit connected to a network and capable of exchanging data with each other. Wherein, of the plurality of computers, one computer inputs the pattern data, the numerical range and the variable step amount of the optical condition parameter, and the input pattern data and optical condition parameter Calculation amount distribution means for obtaining a calculation amount required for the light intensity simulation based on the numerical range and the variable step amount, and distributing the calculation amount to a plurality of computers. , Respectively, distributed by the calculation amount distribution means. The arithmetic processing of the amount of calculation is performed in parallel in each arithmetic processing processor, and the one computer combines the respective arithmetic processing results performed by the other computers except the above, and obtains the light intensity simulation result. Light intensity simulation device. 5. An arithmetic process according to a predetermined calculation algorithm is performed based on pattern data of a photomask pattern and optical condition parameters obtained by parameterizing various optical conditions in an exposure step of the photomask pattern.
In a light intensity simulation method for obtaining a light intensity distribution of a projection optical image when the photomask pattern is exposed and transferred onto a wafer, a first step of inputting the pattern data, a numerical range of the optical condition parameter, and a variable step amount. A second step of obtaining a calculation amount required for light intensity simulation based on the numerical range and variable step amount of the input pattern data and optical condition parameters, and distributing the calculation amount to a plurality of steps; A third stage in which each of the processing processors performs parallel processing of the calculation according to the calculation algorithm for each of the plurality of allocated calculation amounts; and a result of each processing performed in parallel by the plurality of processing processors. And a fourth step of obtaining the light intensity simulation result. Light intensity simulation method. 6. The predetermined calculation algorithm is a numerical integration calculation based on the pattern data and the value of the optical system parameter, wherein the predetermined calculation algorithm changes the optical system parameter by the variable step amount within the numerical value range. In the second step, the numerical value range and the variable step amount of the input optical condition parameters are used to calculate each of the numerical integration results obtained in the second step. Determining the number of combinations of values of the optical condition parameters, distributing the determined number of combinations of the optical system parameters according to the number of the plurality of arithmetic processing processors, and, in the third stage, the plurality of arithmetic processing Each of the processors calculates a combination of the pattern data and the values of various optical system parameters distributed in the second step. Zui, the light intensity simulation method according to claim 5, characterized in that the numerical integration calculation. 7. In the second step, the number of combinations of the values of the various optical system parameters distributed according to the number of the plurality of arithmetic processing processors is determined according to the respective arithmetic processing performances of the plurality of arithmetic processing processors. 7. The light intensity simulation method according to claim 6, wherein the light intensity is changed in accordance with the change. 8. An arithmetic process according to a predetermined calculation algorithm is performed based on pattern data of a photomask pattern and optical condition parameters obtained by parameterizing various optical conditions in an exposure process of the photomask pattern.
A computer-readable recording medium having recorded thereon a light intensity simulation program for obtaining a light intensity distribution of a projection optical image when the photomask pattern is exposed and transferred onto a wafer, wherein the light intensity simulation program is stored in the computer. A first procedure for inputting the pattern data, the numerical range of the optical condition parameter and the variable step amount, and a light intensity simulation based on the input pattern data and the numerical range and the variable step amount of the optical condition parameter A second procedure for obtaining the amount of calculation required for the calculation, and distributing the amount of calculation to a plurality of processors; and for each of the plurality of arithmetic processing processors of the computer, the calculation algorithm assigned to the plurality of processors is followed by the calculation algorithm. A third procedure for performing the arithmetic processing in parallel; A computer-readable recording medium storing a light intensity simulation program for executing the fourth procedure of synthesizing respective arithmetic processing results performed in parallel by the plurality of arithmetic processing processors and obtaining the light intensity simulation result . 9. The predetermined calculation algorithm in the light intensity simulation program is a numerical integration calculation based on the pattern data and the value of the optical system parameter, wherein the optical system parameter is set within the numerical range. Summing and averaging the respective numerical integration calculation results when the variable step amount is changed. In the second step, the numerical integration calculation is performed based on the numerical value range and the variable step amount of the input optical condition parameters. Calculating the number of combinations of the values of the optical condition parameters used in the calculation, and distributing the obtained number of combinations of the optical system parameters according to the number of the plurality of arithmetic processing processors. In the above, the pattern data and the second procedure are stored in each of the plurality of arithmetic processing processors. Based on a combination of the distributed values of the various parameters of the optical system was, and computer readable recording medium the light intensity simulation program according to claim 8, characterized in that to perform the numerical integration calculation. 10. In the second procedure, the number of combinations of the values of the various optical system parameters distributed according to the number of the plurality of arithmetic processing processors is determined based on each arithmetic processing performance of the plurality of arithmetic processing processors. A computer-readable recording medium on which the light intensity simulation program according to claim 9 is changed.