JP2006048271A - Image processing system using grid computing - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To resume an interrupted job from the interrupted point even if a power failure occurs due to an abnormal state of an image processing device in grid computing. <P>SOLUTION: In an image processing system which is composed of a system capable of distributing a load by grid computing (abbreviated as "grid" hereafter) and an image forming device that requests a job to the grid, the image forming device comprises: a job requesting means that divides a job and requests a job to the grid; a data processing means that exchanges data; a job combining means that combines results of the job inputted by the job requesting means; a substitute device list generating means that generates a list of devices capable of performing the same functions as the image forming device; and a job storage means that stores results received from the job even if a power failure occurs during processing. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  In the present invention, a grid computing load distribution system is applied to a printer or a multifunction printer connected to a host computer.

  FIG. 1 is a diagram for explaining the architecture of grid computing applied to the present invention.

  Although there are several types of grids, what is described here is a type called a desktop grid, which executes a job by using the free time of a CPU such as a desktop PC.

  The client in FIG. 1 is a user who submits a job, and the request (job) is transferred to a task manager (hereinafter abbreviated as TM) and the contents are transmitted to a dynamic job scheduler (hereinafter abbreviated as DJS). .

  The DJS performs overall resource management, selects an optimal resource broker, and notifies the TM (this resource refers to a free CPU state here).

  The Broker registers the information of the resource sucked up by the resource manager (hereinafter abbreviated as RM) in the DJS, submits a job to the optimum resource according to the request from the TM, and notifies the TM of the completion.

  The TM submits the job to the optimal Broker selected by the DJS, and thereafter monitors the progress of the job. When the TM receives a completion notification from the Broker, the TM notifies the user of the result. Also, if there is a change / abnormality in the resource (eg, a failure, another job has been accepted, etc.), the RM notifies the Broker.

  The realization of desktop grid computing is that distributed processing is possible by allocating jobs to resources such as CPUs (which are not normally used) that are optimal in such a mechanism.

  Next, the configuration when this technology is applied to the PDL processing of the image forming apparatus will be described with reference to FIG. In the description of FIG. 1, each module constituting the grid is treated as a separate one. However, when applied to a printing apparatus, a plurality of modules generally exist in one device.

  In FIG. 2, a client PC that issues a print instruction (job input) inputs a print job, has an TM and DJS function in the image forming apparatus, and has Broker, RM in PC1, PC2, and PC3, for example, using three PCs. A device configuration to which distributed processing by grid computing is applied.

  A job (PDL print job data) input from a client is divided into resources of PCs 1, 2, and 3 via TM and DJS of the image forming apparatus. At this time, an application program for PDL data image expansion processing is also transmitted from the image forming apparatus at the same time.

The image forming apparatus collects and finally outputs an image that has been subjected to distributed execution of PDL image formation on each PC using a grid mechanism.
This distributed processing target resource may be three or more PCs, and even if it includes the resource of the client PC of the job submission source, the resources of the image forming apparatus on the same network are used as targets. Also good.
IBM professional paper "Applicability of grid computing to commercial systems"<http: // www-6. ibm. com / jp / provision / no36 / pdf / 36_ppr1. pdf>

  In recent years, research on gathering the CPU power of computers using grid computing technology and realizing high-speed processing has been active in science-related research, but application to embedded devices is still unparalleled.

  When this technology is applied to an image forming apparatus, a plurality of image forming apparatuses connected to the LAN environment of the office used can be considered as a grid. Since such an image forming apparatus may be used by a plurality of people in an office, a sudden power interruption or the like may occur in the middle. In such a case, there is a possibility that all the processing that has been performed halfway is wasted. In order to avoid this, a method of transferring control to an alternative server as in JP-A-11-328130 can be considered. However, when data is output by the image forming apparatus, the data is output by another apparatus instead of the apparatus that is desired to be replaced.

  The present invention has been made paying attention to the above points. Grid computing that continues execution of jobs up to that point even when the power is cut off due to an abnormality of an image processing apparatus in grid computing or the like. An object of the present invention is to provide an image processing system to which is applied.

In order to solve the above problems, a system according to the present invention is:
In an image processing system composed of a system capable of load balancing by grid computing (hereinafter abbreviated as “grid”) and an image forming apparatus that requests jobs from the grid.
The image forming apparatus divides the input job, job request means for requesting the job to the grid, data processing means for exchanging data such as image input / output and status output, and the job request means. Job combination processing means for summarizing job results, alternative device list creation means for creating a list of devices that can perform the same functions as image forming devices, and job storage for saving results received from jobs even if the power is cut off during processing Have means.

  In addition, when the grid cannot communicate with the apparatus by means of the apparatus communication means for exchanging data with the image forming apparatus according to the substitute apparatus list created by the substitute apparatus list creating means, It has an alternative device communication means for changing the device list and communicating with the alternative device.

  In addition, the image forming apparatus receives the result of the job requested by the job request unit from the other image forming apparatus and confirms that the requested destination operates, and then transfers the image to the requested image forming apparatus. It has a transfer means.

  When grid computing technology is applied to an image forming apparatus, if the power is cut off during a computing operation and the machine temporarily stops operating, the job can be executed on another machine. If the power is restored and the power can be restored, the information can be collected from other machines and the original image forming apparatus can continue the processing.

  Hereinafter, the best mode for carrying out the present invention will be described in detail with reference to examples.

[First embodiment]
Hereinafter, the present invention will be described in detail with reference to the drawings based on its effective embodiments.

(Hard structure of image forming apparatus)
FIG. 3 is a diagram illustrating an appearance of an image forming apparatus having a copying function. A scanner 201 that is an image input device illuminates an image on paper as a document, and scans a CCD line sensor (not shown) to generate raster image data.

  When the user sets the original paper on the tray 203 of the original feeder 204 and instructs the start of reading in the operation unit 202, the controller CPU of the image forming apparatus gives an instruction to the scanner 201, and the feeder 203 sets the original paper 1 Each sheet is fed, and the scanner 201 performs an original image reading operation.

  An operation unit 202 is a user interface for specifying setting instructions, status display, and various operation settings during a copy operation.

  The printer engine 103, which is an image output device, is a part that prints raster image data on paper. The method includes an electrophotographic method using a photosensitive drum or a photosensitive belt, an ink jet method in which ink is ejected from a micro nozzle array and an image is directly printed on a sheet, and any method may be used. The printing operation is activated by an instruction from the controller CPU.

  The printer engine 103 has a plurality of paper feed stages so that different paper sizes or different paper orientations can be selected, and there are paper cassettes 206, 207, 208 corresponding thereto. The paper discharge tray 205 receives paper that has been printed.

  FIG. 4 is a block diagram illustrating a control configuration of the image forming apparatus. The printer controller 102 is connected to the scanner 201 as an image input device and the printer engine 103 as an image output device. On the other hand, the printer controller 102 inputs and outputs print data, image information, and device information to and from the host and grid via the network 308. do.

  The CPU 301 is a controller that controls the entire system. A RAM 302 is a system work memory used for the CPU 301 to operate. The RAM 302 is also an image memory for temporarily storing image data. A ROM 303 is a boot ROM, and stores a system boot program. An HDD 304 is a hard disk drive and stores system software and image data.

  The operation unit I / F 306 controls an interface with the operation unit (UI) 202 and outputs image data to be displayed on the operation unit 202 to the operation unit 202. Further, it plays a role of transmitting information input by the user via the operation unit 202 to the CPU 301. Further, environment setting information such as an operation mode input from the operation unit 202 is stored in the NVRAM 316 which is a nonvolatile memory.

  The network 308 inputs and outputs information to and from the host 001 and the grid 001a. The above devices are arranged on the system bus 307.

  An image bus interface (Image Bus I / F) 305 is a bus bridge that connects a system bus 307 and an image bus 309 that transfers image data at high speed, and converts a data structure. The following devices are arranged on the image bus 309. A raster image processor (RIP) 310 expands a PDL code transmitted from a network into a bitmap image. A device I / F unit 311 connects the scanner 201 and the printer engine 103, which are image input / output devices, and the printer controller 102, and performs synchronous / asynchronous conversion of image data.

  The scanner image processing unit 312 corrects, processes, and edits input image data. A printer image processing unit 313 performs printer correction, resolution conversion, and the like on print output image data. The image rotation unit 314 rotates image data. The image compression unit 315 performs JPEG compression / new tone processing for multi-valued image data, and performs compression / decompression processing such as JBIG, MMR, and MH for binary image data.

(Software configuration of image forming apparatus)
FIG. 5 is a diagram illustrating an example of use of the image forming apparatus in grid computing according to the present exemplary embodiment. The background grid computing technology and configuration are the same as those described in FIGS.

  Here, the host (501) is a PC that can submit jobs to the server, and the servers (502, 503, 504) mean image forming apparatuses such as MFPs and SFPs, and Task Manager is the same as the printer in FIG. , Dynamic Job Scheduler function. The server has three grids 505, 506, and 507, and the received job can be divided into a plurality of jobs and distributed according to the grid processing speed and the like. The server integrates the distributed jobs and returns the results to the host.

The sequence diagram of FIG. 6 shows the flow of general processing up to the end of the job when a job is submitted from the host to the server in FIG.
First, a job is sent from the host (501) to the server (502) as a job request (1). Here, the server (502) transmits the alternative server list (508) to the host (501) in case the power is cut off in the middle (2). In the alternative server list, the name of the server is a list, and if the server that made the request is temporarily down, an inquiry can be made to the next server in the list. The server analyzes the requested job, divides it into two jobs, and transfers each job to the grids 505 and 506. Here, a request (3) for a job divided into 505 is first performed, and an alternative server list (508) is transmitted in preparation for power failure (4). If there is no response from the server that requested the job, that is, the server at the top of the list, refer to this list and notify other servers of the job completion status or job execution status. . The same operation is performed for 506 (5) and (6), and the server waits for the grid job to end.

  Each grid receives the transmitted job and processes it. When the processing is completed, the top of the alternative server list (508) is viewed, and a job end notification is transferred to the server (502) (7). Here, it is confirmed that the server (502) is operating, and the data output when the job is performed is transferred (8). Similar processing is performed on the other grid (506) ((11), (12)).

  Since the host (501) does not know the operation of the server or the grid operating as described above, the server (502) makes an inquiry to the server (502) via the alternative server list to check the situation at any time ((9), (10), (13), (14)). For example, at the time of the inquiry in (9), the host side is notified that one of the two processes has been notified of the end of one job. In (13), information indicating that all jobs have been completed and the final processing of data is being performed is returned.

  When all requested jobs are returned (13), the server (502) collects the data, prints necessary data, etc., and finally notifies the host (501) of the end of the job (15). ) Job execution is complete.

  Next, the case where the server power is turned off in the configuration of FIG. 5 will be described using the configuration diagram of FIG. 7 and the sequence diagram of FIG.

  In FIG. 8, the processing of the job transferred to the grid (505) is completed (7), and the result data is transmitted (8). Here, it is assumed that the server (502) is powered off. The grid (506) that has finished executing the job transmits a job end notification to the server (502). However, since there is no response, the alternate server list is changed and the job end notification is transmitted to the server (503) ( 10), the alternative server list is also sent to the server, and the processing result data is sent (11).

  Here, the same processing is performed in the host (501). The server (502) is sent to confirm the status (12), but the server (502) is down because there is no reply, so the server (503), which is the next candidate in the alternative server list of the host, is sent. The status is confirmed (13), and the server (503) notifies that the server (502) is currently stopped and the execution result data is obtained from the grid (505) (14). Here, since all the processes transferred to the grid have been completed, the server (503) continues to send a data transmission notification to the server (502) (15).

  Next, when the server (502) returns, it responds to the data transmission (16) of the server (503), and receives the data that has been received by the server (503) (17). Further, since the host can also know that the server (502) has returned from the server (503) (19) when the status confirmation (18) is performed, the server (502) confirms the status again at this point. (20), it becomes possible to confirm the operation of the job (21).

  Finally, the server (502) sends a termination notice to the host (501) after performing a desired operation based on the data received from the grid (505) and the data received via the server (503) ( 22) Complete.

  FIG. 8 shows a data flow different from that in FIG. It is shown that when the server (502) is down, all the alternative server lists are rearranged and data is collected in the server (503).

  According to the operation described above, even if the server is down due to an unintended power cut or the like during the process, the process can be performed by substituting with another server and transferring the data when the server is restored.

It is a figure which shows the structure of the grid computing of background art. It is a figure which shows the structure at the time of applying to PDL processing of background art 1 is a diagram illustrating an overview of an image forming apparatus. 2 is a block diagram illustrating a control configuration of the image forming apparatus. FIG. It is a figure showing the structure of a 1st Example. It is the sequence diagram which illustrated the flow of the data of a 1st Example. It is a figure showing the structure when the power interruption of a 1st Example has occurred in the middle. It is a sequence diagram when the power interruption of the first embodiment occurs in the middle. 001 Host computer 102 Printer controller overall 103 Printer engine

Claims (3)

In an image processing system comprising a load distributable system by grid computing (hereinafter abbreviated as “grid”) and an image forming apparatus that requests a job to the grid.
The image forming apparatus divides a job, requests a job to a grid, a job request unit that exchanges data, a data processing unit that exchanges data, a job combination processing unit that collects job results input by the job request unit, An alternative device list creation unit that creates a list of devices that can perform the same function as an image forming device, and a job storage unit that stores a result received from a job even when the power is turned off during processing. An image processing system using grid computing.   If the grid cannot communicate with the apparatus by the apparatus communication means for exchanging data with the image forming apparatus according to the alternative apparatus list created by the alternative apparatus list creation means, and the apparatus communication means cannot communicate with the apparatus, The image processing system to which grid computing is applied according to claim 1, further comprising: an alternative device communication unit that changes the list and communicates with the alternative device.   2. The image forming apparatus according to claim 1, wherein the grid includes an alternative transfer unit that receives a result of a job requested by a job request unit from another image forming apparatus and transfers the result to the requested image forming apparatus. An image processing system to which the grid computing according to 1 is applied.

Images