JP4479561B2 - Information processing system and information processing method - Google Patents

Description

  The present invention relates to an information processing system, an information processing method, and a computer program for coordinating operations between two or more devices connected to a network, and in particular, using processing capabilities of a plurality of computers connected to a network. The present invention relates to an information processing system, an information processing method, and a computer program that execute an enormous number of arithmetic processes.

  More specifically, the present invention relates to an information processing system and an information processing method, in which a computer including one or more processors is used as a node, and a plurality of sequential and coupled jobs are executed using a plurality of nodes connected to a network. In particular, the present invention relates to an information processing system, an information processing method, and a computer program for obtaining a result of a succeeding job without waiting for the preceding job to be completely completed in the sequential coupled job processing. .

  Along with the recent rapid progress in the information technology (IT) field, various forms of computer systems have been developed and manufactured, and are widely used in universities, other research institutions, corporate offices, and even general households. . On a computer, in addition to text-format document files, various media such as audio, images, and natural languages are digitized and handled mathematically, so that information editing / processing, storage, management, transmission, sharing, etc. are more advanced. It is possible to perform a wide variety of processing.

  More recently, research and development on distributed computing technology that promotes high computing performance by coordinating a plurality of computers on a network and achieving high computing performance is being promoted.

  The present inventors consider that such a distributed computing technique is appropriate for an arithmetic process in which the same calculation is repeatedly performed while changing conditions and parameters. As an example of this type of arithmetic processing, simulation calculation for powder behavior analysis in an image forming apparatus such as a copying machine or a printer using electrophotographic technology can be cited. In this case, a powder consisting of two components, toner as an image forming agent and a carrier made of a magnetic material for transporting the toner, is handled as particles to be analyzed, and powder behavior analysis is performed in each process such as stirring, development, and transfer. By applying simulation, an image to be formed can be predicted and evaluated without actually performing an image formation experiment.

  The simulation calculation of the electrophotographic process is composed of sequentially coupled jobs. For example, in the development process, calculation of developer agitation in the developing unit, formation of magnetic brushes (ears) on the development roll and calculation of charge, calculation of developer flight from the development roll to the photoreceptor, Sequentially coupled.

  Here, it is considered that it is not effective from the viewpoint of the utilization efficiency of the node if the subsequent job is executed after waiting for the completion of the preceding job that is sequentially coupled.

  For example, in the above-described electrophotographic process, in the process from the stirring of the developer to the development, the developer that has been transferred from the auger portion to the magnet roll portion is sequentially developed on the photoconductor. After completing the calculation of the charge amount, start calculating the length of the magnetic brush, and after completing the length calculation of the magnetic brush, calculate the number of particles developed on the photoconductor. When trying to do so, the subsequent processing is not started easily, and the utilization efficiency of the nodes is not good.

  For example, in a distributed system where multiple computers are connected to a network, the output data of a process on one computer is transferred as a process input data on another computer via a pipe. When performing parallel processing, a data partitioning method has been proposed that allocates data in proportion to the capacity of each computer's processing capacity, thereby reducing variations in processing time at each node and improving system utilization efficiency. (For example, refer to Patent Document 1).

  However, this discloses an efficient processing method between nodes for a single job. From the viewpoint of efficient use of processors when executing sequential and coupled jobs. Not necessarily effective.

JP-A-9-185590

  An object of the present invention is to provide an excellent information processing system, information processing method, and computer program capable of suitably linking operations between two or more devices connected to a network.

  A further object of the present invention is to provide excellent information capable of suitably executing a plurality of sequential and coupled jobs using a plurality of nodes connected to a network, with a computer including one or more processors as nodes. A processing system, an information processing method, and a computer program are provided.

  A further object of the present invention is to provide excellent information that can obtain a result of a succeeding job without waiting for the preceding job to be completely completed when processing a sequentially coupled job in a distributed computer environment. A processing system, an information processing method, and a computer program are provided.

  The present invention has been made in consideration of the above problems, and a first aspect of the present invention is that a computer including one or more processors is a node, and a plurality of sequential and coupled jobs are connected to a network. Information processing system using each node, each node can independently execute parallel jobs, and can sequentially execute coupled jobs. The coupled jobs are sequentially executed on the processed data, and the execution result is output to the other node every time the processing of each job is completed. The information processing system is characterized in that the job is executed with respect to the processing target data input to itself based on the job execution result of the above.

  According to the distributed computing technology, it is possible to link a plurality of computers on a network and realize high calculation performance through the cooperative operation, and increase the return on investment.

  However, in a distributed computer environment, when trying to process a plurality of jobs that are successively coupled, if the subsequent job is executed after waiting for the first job to finish completely, the viewpoint of processor utilization efficiency There is a problem that it is no longer effective.

  Therefore, in the present invention, when processing sequentially coupled jobs in a distributed computer environment, a mechanism for obtaining a result of a subsequent job without waiting for the previous job to be completely completed is introduced.

  For example, in a job that is sequentially and coupledly executed as A → B → C → D..., B job calculation is started at another node of the distributed system before A job calculation is completely completed. To do. At this time, part of the data that has been calculated in job A is sent to job B, and the calculation of job B is simultaneously performed using this data. As a result, the job of B can be started and the result of B can be obtained at the same time without waiting for the job of A to finish.

  For example, simulation calculations in the electrophotographic process including exposure / charging, development, transfer, and fixing, calculation of developer agitation in the developing unit, formation of magnetic brush (earing) on the developing roll, and calculation of charging , Each job for calculating the flying amount of the developer from the developing roll to the photosensitive member. Then, if the developer for behavior analysis is divided into a plurality of particle groups and the simulation calculation for each particle group is distributed to each node, the subsequent job can be completed without waiting for the first job to be completed completely. Job results can be obtained. That is, a plurality of simulation calculations for each particle group can be performed simultaneously. For example, the conveyance calculation can be performed in order from the developer transferred from the auger portion to the magnet roll portion.

  According to a second aspect of the present invention, there is provided a computer described in a computer-readable format so that a process for executing a plurality of sequential and coupled jobs using a plurality of nodes is executed on a computer system. A program that causes one of the nodes arranged before and after to execute the above-mentioned coupled job sequentially for the input processing target data, and to execute it each time processing of each job is completed A procedure for outputting the result to the other node; and a procedure for causing the other node to execute the job on the processing target data input to itself based on a job execution result by the one node. This is a computer program characterized by the above.

  The computer program according to the second aspect of the present invention defines a computer program described in a computer-readable format so as to realize predetermined processing on a computer system. In other words, by installing the computer program according to the second aspect of the present invention in the computer system, a cooperative action is exhibited on the computer system, and the information processing according to the first aspect of the present invention is performed. The same effect as the system can be obtained.

  According to the present invention, an excellent information processing that can suitably execute a plurality of sequential and coupled jobs using a plurality of nodes connected to a network using a computer including one or more processors as a node. A system, an information processing method, and a computer program can be provided.

  In addition, according to the present invention, when processing sequentially coupled jobs in a distributed computer environment, it is possible to obtain the result of a subsequent job without waiting for the previous job to be completely completed. An information processing system, an information processing method, and a computer program can be provided.

  Other objects, features, and advantages of the present invention will become apparent from more detailed description based on embodiments of the present invention described later and the accompanying drawings.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  The present invention relates to distributed computing that links operations between two or more devices connected to a network. According to distributed computing, a plurality of computers on a network can be linked together, and high computing performance can be realized by the cooperative operation.

  FIG. 1 schematically shows the configuration of a distributed computing system. On the network, a server that controls arithmetic processing such as simulation calculation and one or more client PCs are connected. In addition to a single LAN segment, the network can be composed of a plurality of LAN segments interconnected via routers or gateways, or a broadband network such as the Internet.

  According to the distributed computing system, it is possible to link a plurality of computers on the network and realize high calculation performance through the cooperative operation, and increase the return on investment.

  For example, a simulation of the behavior of a system constituted by a plurality of processes can be made into a simulation module for each process, and execution of the simulation module can be assigned to each client PC. In the electrophotographic process, the simulation is modularized for each process such as exposure / charging, development, transfer, fixing, etc., and each simulation module is allocated to the client PC, and the server integrates these simulation results to evaluate the entire image forming apparatus. Can be performed.

  However, in a distributed computer environment, when trying to process a plurality of jobs that are successively coupled, if the subsequent job is executed after waiting for the first job to finish completely, the viewpoint of processor utilization efficiency There is a problem that it is no longer effective.

  FIG. 8 shows this state. In order to efficiently process the process, resources (nodes) of the distributed system (cluster) are allocated according to the data processing capacity of the job and the processing capacity of the node. In this case, the job is processed sequentially. When there is a job that is sequentially and continuously executed in the order of A → B → C → D..., The subsequent job is performed after waiting for the preceding job to be completely completed. Accordingly, since the processor that executes the subsequent job is in a standby state until the preceding job is completely completed, the use efficiency is not good.

  For example, when a distributed computer as shown in FIG. 8 is used for the electrophotographic process simulation, the calculation of the length of the magnetic brush is started after the calculation of the charge amount for all the developer particles to be analyzed. However, if it is attempted to calculate the number of particles developed on the photosensitive member after the length calculation of the magnetic brush is completely completed, the subsequent processing is not easily started and the utilization efficiency of the processor is not good.

  Therefore, in the present embodiment, when processing sequentially coupled jobs in a distributed computer environment, a mechanism for obtaining a result of a subsequent job without waiting for the previous job to be completely completed is introduced.

  FIG. 2 shows this state. In a job that is sequentially and coupledly executed as A → B → C → D..., Before the job calculation of A is completely completed, the calculation of the job of B is started at another node of the distributed system. At this time, part of the data that has been calculated in job A is sent to job B, and the calculation of job B is simultaneously performed using this data. As a result, the job of B can be started and the result of B can be obtained at the same time without waiting for the job of A to finish.

  Here, with reference to FIG. 3, a case will be considered in which a plurality of jobs sequentially coupled in the order of JOB1 → JOB2 → JOB3 are distributedly processed using a plurality of nodes # 1 to # 4.

  Each node # 1 to node # 4 can independently execute parallel jobs of JOB1 → JOB2 → JOB3 sequentially in parallel.

  First, when data to be processed is input, the node # 1 executes the first job JOB1 with respect to the data, and when the processing is completed, the subsequent jobs JOB2 and JOB3 are sequentially executed with respect to the execution result. To do.

  Node # 1 outputs the execution result of JOB1 to node # 2. Node # 2 executes job JOB1 on the processing target data input to itself based on the execution result of node # 1. When this processing is completed, the subsequent jobs JOB2 and JOB3 are sequentially executed on the execution result, and the execution result of JOB1 is output to the node # 2.

  Similarly, node # 3 executes job JOB1 on the processing target data input to itself based on the execution result by node # 2.

  Further, when the processing of job JOB2 is completed, node # 1 executes subsequent job 3 following the execution result, and outputs the execution result of job 2 to node # 2. Node # 2 executes job JOB2 on the processing target data input to itself based on the execution result by node # 1. When this processing is completed, subsequent job JOB3 is sequentially executed with respect to the execution result, and the execution result of JOB2 is output to node # 3.

  Similarly, node # 3 executes job JOB2 for the processing target data input to itself based on the execution result by node # 2.

  Below, the case where the distributed computing technique which concerns on this embodiment is applied with respect to the simulation calculation of an electrophotographic process is demonstrated.

  FIG. 4 illustrates the configuration of an electrophotographic process apparatus, focusing on the development process. Around the photoreceptor 13, an intermediate transfer belt 14, a brush roller 34, a charging roller 36 and a developing unit 12 are provided in order along the rotation direction b. The intermediate transfer belt 14, the brush roller 34, and the charging roller 36 are all in contact with the photosensitive surface of the photoreceptor 13. Further, between the charging roller 36 and the developing unit 12, an LED array head 40 that performs line exposure on the photosensitive surface is disposed.

  The developing unit includes a developing roller 38 disposed so as to face the photosensitive member 13, screw feeders 39 </ b> A and 39 </ b> B that are positioned below the developing roller 38 and supply a two-component developer to the developing roller 38, A developing roller 38 and a housing 37 for accommodating the screw feeders 39A and 39B are provided. The two-component developer contains toner and magnetic carrier particles as main components. An opening 37 </ b> A is provided in a portion of the housing 37 that faces the photoreceptor 13.

  The developing roller 38 is disposed so that a gap, that is, a developing gap is formed between the photosensitive roller 13 and the photosensitive surface. The developing roller 38 includes a cylindrical magnet roll 38B and a sleeve 38A that covers the magnet roll 38B. The magnet roll 38B has a cylindrical shape and is fixed to the main body of the image forming apparatus. The sleeve 38A has the same counterclockwise direction as the rotation direction b of the photoconductor 13 around the axis of the magnet roll 38B, that is, the sleeve 38A. It rotates in the opposite direction relative to the photoconductor 13 at the portion facing the photoconductor 13. Thereby, the transfer efficiency of the toner from the developing roller 38 to the photosensitive member 13 is enhanced.

  The magnet roll 38B is a mag roller in which magnetic material powder such as ferrite or rare earth magnet alloy is formed into a columnar shape or a cylindrical shape, and is magnetized so that the N pole and the S pole are arranged in a predetermined pattern. It is formed by sintering while. As the magnetized pattern, a portion facing the photosensitive member 13 is the developing pole S1, and a pick-off pole N1 is located next to the developing pole S1 along the rotation direction of the sleeve 38A, and a pickup pole N2 and a trimming pole are adjacent to the pick-up pole N1. Examples include a pattern in which magnetic poles are arranged in the order of S2 and the transport pole N3. The development pole S1 and the trimming pole S2 are all S poles, and the pickoff pole N1, the pickup pole N2, and the transport pole N3 are all N poles.

  The portion corresponding to the developing nip in the developing pole S1 is magnetized so that the change in the normal direction magnetic flux density Br becomes ± 5 [mT]. Here, assuming that the boundary between the transport pole N3 and the trimming pole S2 is 0 degree and the clockwise direction is a positive angle, the development nip is located at a part of 240 to 270 degrees and includes the development nip part. Thus, the development pole S1 is formed. The change in the normal magnetic flux density Br is ± 5 [mT] at the angle of 240 to 270 degrees in the magnet roll 38B. This indicates that the change in the normal magnetic flux density Br at the developing nip portion of the magnet roll 38B is magnetized so as to be ± 5 [mT]. The magnet roll 38B is further magnetized so that the minimum value of the normal magnetic restraining force Fr is located in the developing nip.

  A trimming blade 41 that aligns the height of the magnetic brush in cooperation with the trimming pole S <b> 2 extends toward the developing roller 38 at a portion facing the trimming pole S <b> 2. A negative developing bias voltage is applied to the developing roller 38.

  In the developing mode, the photosensitive member 13 rotates counterclockwise at a constant speed, so that the photosensitive surface is negatively charged by the charging roller 36. Next, the charged surface of the photosensitive surface is exposed by the LED array head 40, whereby the potential of the exposed portion of the charged surface is lowered to form an electrostatic latent image. In the developing unit 12, the toner charged to a negative voltage is electrically attached to the electrostatic latent image formed on the photosensitive surface, that is, the potential lowering portion of the charging surface by the developing roller 38 as in the photosensitive member 13. Development is performed to form a toner image. The toner adhering to the photosensitive surface is electrically drawn toward the intermediate transfer belt 14 by the transfer roller 32 to which a positive transfer voltage having a polarity opposite to that of the toner is applied. As a result, the toner image on the photosensitive surface 13A is transferred from the photosensitive member 13 to the intermediate transfer belt.

  When the sleeve 38A rotates, the developer supplied into the housing 37 by the screw feeders 39A and 39B is adsorbed on the surface of the sleeve 38A by the pickup pole N2. Here, on the surface of the sleeve 38A, there are a magnetic field from the transport pole N3 toward the development pole S1, a magnetic field from the pick-off pole N1 toward the development pole S1, a magnetic field in the direction from the pickup pole N2 toward the trimming pole S2, and from the transport pole N3. A magnetic field directed to the trimming pole S2 is formed, and the developer has a structure in which toner adheres to the surface of the magnetic carrier particles. Therefore, as shown in FIG. 5, the developer adsorbed on the surface of the sleeve 38A is arranged in the direction of the lines of magnetic force on the surface of the sleeve 38A and rises to form a magnetic brush.

  As shown by the arrow in FIG. 5, the magnetic brush formed on the surface of the sleeve 38A in the vicinity of the pickup pole N2 is trimmed pole S2, transport pole N3, developing pole S1, and pick-off as the sleeve 38A rotates. It is conveyed from the right side to the left side of the page toward the pole N1. Then, the height of the magnetic brush is adjusted when passing through the trimming pole S2, and the toner on the magnetic brush is transferred to the photosensitive member 13 in the vicinity of the developing pole S1, so that the surface of the sleeve 38A is almost only a magnetic carrier. The magnetic brush remains. As the sleeve 38A rotates, the magnetic brush that has almost only the magnetic carrier drops off from the surface of the sleeve 38A at the pick-off pole N1 and returns to the inside of the housing 37.

  In the developing mode, the sleeve 38A rotates in this way, so that fresh developer is always replenished at the pickup pole N2 and conveyed to the developing pole S1, and the toner is transferred to the photosensitive member 13 at the developing pole S1 to be photosensitive. The latent image on the surface 13A is developed.

  The process leading to the agitation and development of the developer is performed by using, for example, an individual element method, and various forces acting on all particles (for example, acting force due to contact such as elastic force and viscous force, van der Worth force). Since an equation of motion is established based on the external force such as the image force, the mirror image force, and the liquid bridge force, and the behavior of each particle is analyzed, a more realistic evaluation can be performed. (For the individual element method itself, for example, the Society of Powder Technology “Introduction to Powder Simulation-Creating Powder Technology with Computers” (Sangyo Tosho Co., Ltd., March 30, 1998), Chapter 3 “Particle Elements” Method simulation ")

  On the other hand, the individual element method has a problem in that the amount of calculation increases when the number of particles handled increases. Therefore, by applying the distributed computing technology according to the present invention and allocating the calculation of each particle to a plurality of nodes, it is considered to realize high computing performance by the cooperative operation of each node and increase the return on investment. .

  Simulation calculation of the process leading to agitation and development includes calculation of developer agitation in the developing unit, formation of magnetic brush (ears) on the development roll and calculation of charge, developer from the development roll to the photoreceptor These flight calculations are sequential and coupled. In the simulation of the electrophotographic process, the particle behavior is calculated by the flow of stirring the developer by the auger and transporting the developer by the magnet roll. As shown in FIG. 6, from the auger part to the magnet roll part. The transport calculation can be performed in order from the developer that has been advected.

  For the particles, the charge amount can be determined by calculating the agitation in the developing device (referred to as JOB1). By calculating the amount of spikes and charging on the developing roll (referred to as JOB2), the length of the spikes can be obtained. Then, the number of particles developed on the photoconductor is determined by performing a flight calculation (referred to as JOB3) from the developing roll to the photoconductor for each particle adhering to the head.

  Each node # 1 to node # 4 can obtain the number of particles to be developed by sequentially executing a coupled job of JOB1 to JOB3 for the particle group distributed to itself.

  Here, paying attention to the node # 1 and the node # 2 arranged before and after, the node # 2 assigns itself to the node # 1 if it does not obtain a calculation result for the particle group previously input to the node # 1. The calculation of the given particle group cannot be started. However, it is not necessary to wait until the flight calculation (JOB3) is completed at the node # 1, and if the result of the agitation calculation (JOB1) is obtained, the agitation calculation for the next particle group can be started.

  Therefore, the node # 1 executes the job JOB1 for the particle group allocated to itself, and when the processing is completed, the node # 1 starts executing the next job JOB2 coupled based on the execution result of the job, A part of the execution result is output to the node # 2. Then, node # 2 executes the same job for the particle group assigned to itself based on the execution result of job JOB1 by node # 1.

  FIG. 7 illustrates a state in which a plurality of jobs JOB1 to JOB3 that are sequentially and coupled are simultaneously executed on a plurality of nodes.

Node # A 1, assigned or particles until particles No. 1 to N _1, the node # 2 is assigned or particles until particles numbers N ₁ + 1 to N _2, the node # to 3 particle numbers N ₂ Particle groups from +1 to N ₃ are assigned. The size of the particle group is, for example, a lump of developer that is transferred from the auger portion to the magnet roll portion.

Node # 1, with respect to particles 1 to N ₁ assigned thereto by performing a stirring calculations, the the process is completed at time t1, based on the calculated charged amount, the particles 1 to N ₁ Next, the calculation of heading and charging is performed. Then, when this calculation is performed at time t2, flight calculation is performed on the same particle group ₁ to N ₁ based on the calculated heading length, and the number of particles developed on the photoconductor at time t3 is calculated. it can. A part of the execution result is output to the node # 2.

Further, when node # 2 receives the charge amount for particle group 1 to N ₁ previously input from node # 1, node # 2 performs agitation calculation for subsequent particle group N ₁ +1 to N ₂ that has been input to itself. The process can be completed at time t2. Then, based on the calculated charge amount and the heading length for the _first particle group 1 to N ₁ obtained from the node # 1, the heading is performed for the particle group N ₁ +1 to N ₂ allocated to itself. The charge calculation is subsequently performed, and this calculation is time t3. Then, based on the calculated heading length and the number of developing particles for the _first particle group 1 to N ₁ obtained from the node # 1, the flight calculation is performed for the particle group N ₁ +1 to N ₂ allocated to itself. And the number of particles developed on the photoconductor at time t4 can be calculated.

  In this way, the conveyance calculation can be performed in order from the developer that has been advected from the auger portion to the magnet roll portion. In other words, a plurality of nodes perform sequential and coupled job processing simultaneously, so that the result of the subsequent job can be obtained without waiting for the first job to be completed. In addition, the utilization efficiency of the Proset node can be increased.

  The present invention has been described in detail above with reference to specific embodiments. However, it is obvious that those skilled in the art can make modifications and substitutions of the embodiment without departing from the gist of the present invention.

  Although the present specification has been described using the embodiment for executing the simulation calculation of the electrophotographic process, the gist of the present invention is not limited to this. Of course, even when the simulation processing other than the electrophotographic process or the processing of sequential and coupled jobs other than the simulation calculation is executed in the distributed computing system, the general sequential output in which the output is output as a flow. The present invention can be similarly applied to job processing.

  In short, the present invention has been disclosed in the form of exemplification, and the description of the present specification should not be interpreted in a limited manner. In order to determine the gist of the present invention, the claims should be taken into consideration.

FIG. 1 is a diagram schematically showing the configuration of a distributed computing system. FIG. 2 is a diagram showing a state in which sequential and coupled jobs are processed in the distributed computer environment according to the present invention. FIG. 3 is a diagram illustrating a state in which a plurality of jobs sequentially coupled in the order of JOB1 → JOB2 → JOB3 are distributed using a plurality of nodes # 1 to # 4. FIG. 4 is a diagram showing the configuration of the electrophotographic process apparatus, focusing on the development process. FIG. 5 is a diagram showing how the magnetic brush is formed on the developing roll. FIG. 6 is a diagram for explaining a simulation calculation of a process leading to stirring and development. FIG. 7 is a diagram showing a state in which a plurality of jobs JOB1 to JOB3 that are sequentially and coupled are simultaneously executed on a plurality of nodes. FIG. 8 is a diagram illustrating a state in which a succeeding job is performed after waiting for the first job to be completed for a plurality of jobs sequentially coupled in a distributed computer environment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 12 ... Developing unit 13 ... Photoconductor 14 ... Intermediate transfer belt 32 ... Transfer roller 34 ... Brush roller 36 ... Charging roller 37 ... Housing 38 ... Developing roller 39 ... Screw feeder 40 ... LED array head 41 ... Trimming blade

Claims (2)

An information processing system that uses a computer including one or more processors as a node, and executes a plurality of sequential and coupled jobs using a plurality of nodes connected to a network,
Simulation calculation in the electrophotographic process including exposure / charging, development, transfer, and fixing, calculation of developer agitation in the developing unit, formation of magnetic brush on the developing roll and calculation of charging, development It consists of a plurality of coupled jobs such as the calculation of developer flight from the roll to the photoconductor. The developer for behavior analysis is divided into a plurality of particle groups, and the simulation calculation for each particle group is performed. Distribute to each node,
Each node receives the data of the particle group to be processed, and the above-described coupled jobs such as developer agitation calculation, magnetic brush formation and charge calculation, and developer flight calculation. Are executed independently and in parallel with other nodes,
One of the nodes arranged before and after sequentially executes the above-mentioned coupled job for the input particle group data, and the execution result of each job is succeeded every time the processing of each job is completed. Output to the other distributed node,
The other node executes the coupled job on the data of the particle group input to the node based on the job execution result by the one node.
An information processing system characterized by this. An information processing method for executing a plurality of sequential and coupled jobs using a plurality of nodes,
Simulation calculation in the electrophotographic process including exposure / charging, development, transfer, and fixing, calculation of developer agitation in the developing unit, formation of magnetic brush on the developing roll and calculation of charging, development It consists of a plurality of coupled jobs such as the calculation of developer flight from the roll to the photoconductor. The developer for behavior analysis is divided into a plurality of particle groups, and the simulation calculation for each particle group is performed. A distribution step to distribute to each node;
Each node, when the data of the particle group to be processed is input, a plurality of the above-mentioned coupled jobs such as developer agitation calculation, magnetic brush formation and charge calculation, and developer flight calculation. An execution step for executing the process in parallel with the other nodes independently of each other;
Have
In the execution step,
One of the nodes arranged before and after sequentially executes the above-mentioned coupled job on the input particle group data, and the execution result of each job is succeeded each time the processing of each job is completed. Output to the other distributed node,
The other node executes the coupled job on the data of the particle group input to the node based on the job execution result by the one node.
An information processing method characterized by the above.

Images