JP5064336B2 - Image forming system and program - Google Patents

Description

  The present invention relates to an image forming system and a program.

  The distributed printing system disclosed in Patent Document 1 divides one print job into a plurality of distributed jobs and prints using a plurality of printers connected to a network. At this time, attribute information of the print job is detected. Based on the detected attribute of the print job, when copy printing is specified for the print job, the print job is divided and set as copy printing. If not, the page is divided into pages.

  In addition, the image forming system disclosed in Patent Document 2 distributes each image forming job obtained by dividing data in units of pages to each printer in a group constituted by one or more printers, and executes each image forming job. By dividing the image forming job so that the image forming jobs distributed to the printers are almost completed at the same time, the printing time when the print data is divided and distributed to a plurality of printers is shortened.

JP 2006-024005 A JP 2006-17884A

  However, since the distributed printing system disclosed in Patent Document 1 controls print jobs via a network, it is impossible to transfer a large amount of image data developed for printing at high speed. For this reason, the distribution of print jobs must be transferred in an intermediate language state, and a print controller for each printer is required separately. That is, a plurality of independent printers are connected, and there is a problem that the cost is not reduced for a high-speed printer system. In addition, the performance of the entire system, such as when a failure occurs in a part of a distributed job, when the printing speed varies, or when you want to submit a print job that has priority over a print job that has been printed in advance. Has the problem of not going up.

  Therefore, it is conceivable to apply the conventional print job distribution technology via a network to image data communication via a high-speed serial transmission path such as PCI Express, but until data is transmitted in response to a data transfer request. There is a problem that an intended data transfer band cannot be obtained due to the influence of latency (delay time). For example, in the image forming system disclosed in Patent Document 2, even if jobs are distributed to a plurality of printing apparatuses according to the performance ratio of each printing apparatus, the plurality of printing apparatuses compete for one serial transfer path. Therefore, since the latency required for data transfer increases, the desired performance cannot be obtained in each printing apparatus, and the overall print job processing performance deteriorates.

  The present invention has been made in view of the above, and an object of the present invention is to provide an image forming system and program capable of improving the performance of the entire print job distribution processing via a network.

  In order to solve the above-described problems and achieve the object, an image forming system of the present invention transfers data between a memory storing image data and the memory without using a CPU that controls the entire system. A single or a plurality of image forming means for forming an image of the image data stored in the memory, and connected to the image forming means via a high-speed serial transmission line; The image data relating to the print job submitted via the network is stored in the memory, and the image data requested by the request issued by the DMAC is transferred from the memory to the requesting image forming unit. The image forming means returns the image data in response to a predetermined request from the control means. The split number of requests, which is the number of subsequent requests issued continuously, is used as the latency of the request, which is the delay time until receiving the image data for the request in the state where the image forming means is operating. Change accordingly.

  The program of the present invention includes a direct memory access controller (DMAC) that transfers data to and from a memory that stores image data without using a CPU that controls the entire system. The program is stored in the memory. The image data relating to a print job that is connected via a high-speed serial transmission path to one or a plurality of image forming units that perform image formation of the image data that has been received is stored in the memory And a program for causing the image data requested by the request issued by the DMAC to function as a control unit for transferring the image data from the memory to the requesting image forming unit. The image data corresponding to the request is subsequently issued while returning from the control means The request split number, which is the number of requests, is changed according to the latency of the request, which is the delay time until receiving the image data for the request in a state where the image forming means is operating.

  According to the present invention, the number of splits when each image forming unit issues a request for image data to the control unit is changed according to the latency (delay time) of the request required for image data transfer (for example, latency). The greater the number of requests, the greater the number of request splits, and the smaller the latency, the smaller the number of request splits). Therefore, the performance of the entire print job distribution processing via the network can be improved.

  Exemplary embodiments of an image forming system and a program according to the present invention are explained in detail below with reference to the accompanying drawings.

[First Embodiment]
A first embodiment of the present invention will be described with reference to FIGS. FIG. 1 is a block diagram showing the configuration of the image forming system according to the first embodiment of the present invention. The image forming system shown in FIG. 1 connects four printing apparatuses A, B, C, and D via a serial transmission path 2 and a switch 3 to one printing controller 1 that functions as a control unit. is doing. As the serial transmission path, a high-speed serial interface (PCI Express (hereinafter referred to as PCIe)) capable of transferring a large amount of print data at high speed is used. Therefore, the switch 3 of the present embodiment is a PCIe switch. As described above, a large number of printing apparatuses A, B, C, and D are connected to one printing controller 1 via the serial transmission path 2 and the switch 3 using PCIe. Print jobs can be printed at high speed and low cost.

  Here, the PCIe connection port of the print controller 1 is RootComplex 4, and the PCIe connection ports of the printing apparatuses A, B, C, and D are Endpoint5.

  The print controller 1 is mounted with a CPU 6 that executes various controls such as print job control. The CPU 6 is connected to an HDD (Hard Disk Drive) 8 via the HDD I / F 7 and is connected to the memory 10 via the memory I / F 9. The CPU 6 is connected to the network 12 via the network I / F 11. Note that the HDD 8 may be replaced with another recording device (for example, a memory card equipped with a flash memory) as long as the recording device has a capacity capable of recording print data of a large amount of print jobs.

  Next, a basic operation for executing one print job with one printing apparatus (for example, printing apparatus A) will be described.

  FIG. 2 shows a data flow from when a print job is input to the print controller 1 until image data for printing is stored in the HDD 8. As shown in FIG. 2, first, the print job is transmitted to the print controller 1 via the network 12. Here, it is assumed that the print job is print language data having a smaller capacity than the image data. The print language is converted into image data for printing by a development process by software of the CPU 6 and is written in the memory 10 in units of one page. The print data written in the memory 10 is sequentially transferred to the HDD 8 and then deleted. When all the development processes of the CPU 6 are completed, the image data of all pages included in the print job is recorded in the HDD 8.

  FIG. 3 shows the flow of data during printing. As a schematic flow at the time of printing, the print controller 1 transfers print data in the HDD 8 page by page to the RootComplex 4 of the PCIe, and the print data is transferred from the RootComplex 4 to the printing apparatus A via the PCIe switch 3. And printed on paper. This will be described in detail below.

  As shown in FIG. 3, the printing apparatus A (B, C, D) transfers data between, for example, an electrophotographic plotter unit 21 that functions as an image forming unit and the memory 10 without using the CPU 6. A direct memory access controller (DMAC) 22, a latency detector 23, and a split issue controller 24.

  As shown in FIG. 3, the print controller 1 at the time of printing transmits print instruction information to the plotter unit 21 of the printing apparatus A by a memory write packet passing through the PCIe 2 and the PCIe switch 3 (shown in FIG. 3 ( 1)).

  The plotter unit 21 that has received the print instruction transmits an activation signal to the DMAC 22. Upon receiving the activation signal, the DMAC 22 requests the print controller 1 to transfer image data via the PCIe Endpoint 5 by a memory read request ((2) shown in FIG. 3). In addition, the DMAC 22 notifies the latency detection unit 23 of the time information when the Read request is issued. Since PCIe2 processes serial packet transfer, the DMAC 22 divides image data constituting the entire print job into units of small size, for example, 128 bytes, and issues transfer requests as a plurality of memory read requests.

  The Read request reaches the HDD 8 that stores the print data via the RootComplex 4 of the print controller 1. Then, print data for the requested Read request is transferred from the HDD 8 to the RootComplex 4. The print data transferred to the RootComplex 4 is transferred to the DMAC 22 via the PCIe 2 and the PCIe switch 3.

  When the DMAC 22 receives the Read data, the DMAC 22 notifies the latency detection unit 23 of reception time information until the Read data received from the DMAC 22 is transferred. The DMAC 22 further transfers the received read data as image data to the plotter unit 21 and sequentially prints the image data on paper ((3) shown in FIG. 3).

  The latency detection unit 23 calculates the latency (delay time) from the request issuance to the data reception from the time information when the Read request received from the DMAC 22 is issued and the reception time information until the Read data received from the DMAC 22 is transferred. ) Is detected and notified to the split issue control unit 24. Here, the latency (or simply latency) of the Read request refers to the time taken until the Read data for the Read request arrives.

  The split issuance control unit 24 notifies the DMAC 22 of the number of split issuances when issuing a read request in accordance with the latency received from the latency detection unit 23. Details of the operation of the split issue control unit 24 will be described later.

  The DMAC 22 determines the number of splits when issuing a Read request in accordance with the number of issued splits. Here, the number of split requests issued (or the number of splits) refers to the number of subsequent Read requests that are issued before Read data for a certain Read request returns from the request destination.

  Here, the relationship among the number of splits, the Read request, and the transfer rate of Read data to the printing apparatus will be described with reference to the timing charts of FIGS.

  FIG. 4 is a timing chart showing an operation example when the number of splits is “1”. As shown in FIG. 4, when the number of splits is “1”, after issuing one Read request 1, the next Read request 2 is not issued until the Read data D1 is received. For this reason, when latency occurs, as shown in the timing chart of FIG. 4, there is an interval in receiving packets of Read data D1, D2, D3..., And the data transfer band is narrowed. In the example shown in FIG. 4, the transfer rate is reduced to about half.

  FIG. 5 shows an example in which the number of splits is increased to “2”. When the number of splits is “2”, two requests 1 and 2 are issued before the first read data D1 is transferred from the printing apparatus. For this reason, the packets of the read data D1 and D2 are transferred without an interval. When the read data D1 is received, the next read request 3 is issued. Therefore, the read data after the read data D2 and D3 are also transferred without any interval, and the maximum effective bandwidth can be obtained. . In the example shown in FIG. 5, data transfer is performed with an efficiency of 100%.

  FIG. 6 is a timing chart showing an example when the number of splits is “2” and the latency is further increased. Since the latency increases, the interval between the read request 2 and the read request 3 from the printing apparatus widens, so that a gap is generated in the interval between the received read data D2 and the read data D3. For this reason, the transfer band when the latency increases may deteriorate to about 2/3.

  Therefore, when the latency increases, as shown in the example shown in FIG. 7, if the number of splits is increased to “3”, the Read data packet can be received without any gap, and the maximum transfer efficiency can be obtained.

  Therefore, the split issuance control unit 24 calculates the number of splits of the Read request issued by the DMAC 22 according to the latency so that the number of splits increases as the latency increases and the number of splits decreases as the latency decreases. .

  As described above, according to the present embodiment, each printing apparatus uses the number of splits when each plotter unit 21 issues a request for image data to the print controller 1 to determine the latency (delay) of the request required for image data transfer. The number of request splits is increased as the latency is increased (for example, the request split number is decreased as the latency is decreased), so that distributed processing is performed by the plurality of plotter units 21 to increase the latency. Even in this case, since the image data can be transferred at the maximum data transfer rate, the performance of the entire print job distribution process via the network can be improved.

[Second Embodiment]
Next, a second embodiment of the present invention will be described with reference to FIGS. The same parts as those in the first embodiment described above are denoted by the same reference numerals, and description thereof is also omitted.

  In the image forming system described in the second embodiment, the smaller the latency, the more the number of splits is reduced. When a plurality of printing apparatuses operate simultaneously, a printing apparatus with low printing performance issues an excessive number of read requests. This is to prevent issuing to the print controller 1. However, if the number of splits of the read request of the slow printing device is too large, a large number of read requests from the slow printing device are queued in the internal buffer of the print controller 1 or PCIe 2 until the transfer of the read data is completed. The buffer will not be released. For this reason, there arises a problem that the read data transfer rate from other printing apparatuses having high printing performance is lowered.

  Therefore, in the present embodiment, the split issue control unit 24 sets the number of read requests issued by the DMAC 22 according to the data transfer bandwidth required for the maximum printing speed of each printing apparatus, so that the printing performance is high. The number of splits is calculated so that the number of splits increases as the maximum data transfer bandwidth required for formation increases, and the number of splits decreases as the printing performance decreases (the maximum data transfer bandwidth required for image formation decreases). is there.

  Here, a specific example of the split number calculation value of the split issuance control unit 24 will be described with reference to FIGS.

  FIG. 8 shows the relationship between the latency and the number of splits when the data size of the read request is 128 bytes and the plotter unit 21 of the printing apparatus exhibits maximum performance at a data transfer rate of 1 GB / s. FIG. 9 shows the relationship between the maximum data transfer bandwidth of the plotter unit 21 of the printing apparatus and the number of splits when the latency is 1024 ns and the data size of the read request is 128 bytes.

The split number is calculated by the split issue control unit 24 from the latency of Read data for the Read request, the payload size per packet, the maximum bandwidth of the plotter unit 21 and the maximum bandwidth of PCIe2 (data transfer path). It can be derived by processing.
Number of splits = (latency (ns) / (1-byte transfer time (ns) × payload size (bytes)) * plotter band (GB / s) / PCIe band (GB / s)

here,
1 byte transfer time (ns) = 1 / PCIe band (GB / s)
It is.

  In addition, since the minimum number of splits is 1, when the calculation result is smaller than 1, the number of splits may be set to 1. If the calculation result is not an integer, the number of splits may be determined by rounding up.

Here, the following conditions are given as specific examples.
PCIe band = 1 GB / s (equivalent to PCIe standard 4-lane connection. One lane is 250 MB / s in one direction)
Plotter band = 1 GB / s
Payload size of Read data = 128 bytes
Latency = 1024 ns
For the above conditions,
Number of splits = 1024 (ns) / (1/1 (GB / s) * 128 (bytes)) * 1 (GB / s) / 1 (GB / s)
= 8.

  Through the above calculation, the split issuing control unit 24 notifies the DMAC 22 of a signal for instructing the number of splits “8”.

  Here, the relationship between the number of splits derived by the above calculation, the latency, and the bandwidth (performance) of the plotter unit 21 is shown in FIG. As shown in FIG. 10, by controlling the number of splits, it is possible to transfer Read data from the print controller at a data transfer rate at which the maximum performance of each plotter unit 21 can be exhibited.

  As described above, changing the number of splits according to latency information is particularly effective when distributed printing processing is performed by a plurality of printing apparatuses. This is because the latency changes depending on the number of printing apparatuses (plotter units 21) activated simultaneously. As in the example illustrated in FIG. 11, the latency for the Read request increases as the number of activations of the printing apparatus (plotter unit 21) increases. This is because, as shown in FIG. 12, when the other printing apparatus (plotter unit 21) is activated, the transfer of Read data is different from the case where only one printing apparatus (plotter unit 21) is activated. This is because the time waiting for the transfer of the read data of the apparatus (plotter unit 21) is added to the latency.

  As described above, according to the present embodiment, the number of splits when each plotter unit 21 issues a request for image data to the print controller 1 depends on the latency (delay time) of the request required for image data transfer. By changing (for example, increasing the number of splits as the maximum data transfer band for image formation is wider and decreasing the number of splits as the maximum data transfer band is narrower), the plurality of plotter units 21 performs distributed processing to increase latency. Even in this case, since the image data can be transferred at the maximum data transfer rate, the performance of the entire print job distribution process via the network can be improved.

  In the case of simplifying the configuration, the number of splits for the printing apparatus may be designated directly from the print controller 1 when the plotter unit 21 is started based on the estimated latency value in advance. The example shown in FIG. 13 is an example in which the printing apparatus is not provided with a latency detection unit and split control, and the number of splits is simultaneously notified to the plotter unit 21 when a printing instruction ((1) shown in FIG. 13) is given. . In this case, if there is no change in the number of plotters to be activated and the latency value until the print job is completed, the printing process can be executed at high speed. However, when a print job is input irregularly and the latency and the number of starter plotters change irregularly, it is desirable to provide a latency detection unit 23 and a split issue control unit 24 as shown in FIG.

[Third Embodiment]
Next, a third embodiment of the present invention will be described with reference to FIGS. In addition, the same part as 1st Embodiment mentioned above or 2nd Embodiment is shown with the same code | symbol, and description is also abbreviate | omitted. In other words, the printing apparatuses A, B, C, and D according to the third embodiment of the present invention are also characterized by using the one that controls the number of splits of the Read request according to the latency and the plotter performance as described above. To do.

  FIG. 14 shows a data flow in the image forming system according to the third embodiment of the present invention. In FIG. 14, only the read data transfer from the print controller 1 is shown to simplify the drawing, the print instruction from the print controller 1 to the printing apparatuses A, B, C, and D, and the printing apparatuses A and B. , C, D from the flow of the read request to the print controller 1 is omitted.

  As shown in FIG. 14, the print controller 1 according to the present embodiment divides an input print job into a plurality of printing apparatuses A, B, C, and D, and executes print control. When the print job is composed of a plurality of pages, one print job is divided into a plurality of printing apparatuses A, B, C, and D to perform print control. More specifically, the print controller 1 divides the printing apparatus into groups each having two sets, designates a printing range including a plurality of pages for each group, and further sets the first printing apparatus in each group as the first printing apparatus. Control is performed to print in order from the first page of the designated print range, and the second printing apparatus executes control to print in reverse order from the last page.

  A specific example will be described with reference to the flowchart of FIG. These operations are processed by software in the CPU 6 of the print controller 1 or implemented by hardware logic that performs equivalent operations. Here, in order to simplify the description, it is assumed that the print job is composed of a 20-page document. The operation from when a print job is input to the print controller 1 via the network 12 until print data of all pages is recorded in the HDD 8 is as described above.

  First, when print data of all pages is recorded in the HDD 8 (Yes in steps S1 to S5), the print controller 1 changes the printing devices A, B, C, and D connected to the PCIe switch 3 to the printing devices A, If it is constituted by B, it is grouped into a second group constituted by the printing apparatuses C and D (step S6).

  Next, the print controller 1 determines the page range to be printed by each group based on the performance of each printing apparatus (step S7). In the example shown in FIG. 14, assuming that the performances of the printing apparatuses A, B, C, and D are all the same, 10 pages of 20 pages of jobs are equally allocated to the first group and the second group. (That is, 1 to 10 pages are allocated to the first group, and 11 to 20 pages are allocated to the second group). If the performance of each printing apparatus is not uniform, it is desirable to assign a grouping and a print page range according to the speed. As an example, two printers are grouped in descending order of performance, and a wide print page range is assigned to high performance groups according to the total performance ratio of the printing devices in each group, while narrow printing is performed for low performance groups. By assigning the page range, the time for completing the printing of the entire print job can be shortened.

  After the printing device grouping and page range assignment are completed, the print controller 1 executes a process of transferring print data from the HDD 8 to each of the printing devices A, B, C, and D (step S8).

  Here, print data is transferred in ascending order from the first page of the assigned print page range to the first printing device in each group, and the first printing device is transferred to the second printing device. On the contrary, control is executed so that print data is transferred in reverse order from the last page in the assigned print page range. In the example shown in FIG. 14, print data is transferred to the first group of printing apparatuses A in the order 1, 2, 3,..., And the first group of printing apparatuses B is 10, 9, 8,. The print data is transferred in this order. On the other hand, the print data is transferred to the second group printing apparatus C in the order of 11, 12, 13,..., And the print data is transferred to the second group printing apparatus D in the order of 20, 19, 18. Will be transferred.

  When the printing apparatuses A, B, C, and D start receiving print data, the printing apparatuses sequentially execute print processing from the received page. However, each time the print data of one page is received, the page that has been printed. The page number is notified from the PCIe Endpoint 5 to the print controller 1 via the PCIe switch 3.

  The print controller 1 receives the page number notified from each printing device A, B, C, D while transferring the print data to each printing device A, B, C, D (step S9). If there is a group in which a state in which the first and second page numbers match on the next page is detected (Yes in step S10), transmission of print data to the group is completed (step S11). . In each group, two printers are printing from the first page and the last page respectively, so if the above condition is detected, all pages assigned to that group have been printed. become. With the above operation, when the printing process is advanced and the printing process in all groups is completed (Yes in step S12), that is, the printing of all pages of the input print job is completed, and the entire process is finished. (Step S13).

  As described above, according to the present embodiment, when a print job composed of a plurality of pages is divided into a plurality of printing apparatuses and printing control is executed, the printing apparatuses are grouped into two groups. The printing range consisting of a plurality of pages is designated for each group, and the first printing device of each group prints in ascending order from the first page of the designated printing range, and the second printing device Since printing is performed in reverse order from the last page, even if the performance of each printing apparatus is different, printing can be completed at the maximum speed without always stopping the two printing apparatuses.

[Fourth Embodiment]
Next, a fourth embodiment of the present invention will be described with reference to FIGS. The same parts as those in the first to third embodiments described above are denoted by the same reference numerals, and description thereof is also omitted. That is, the printing apparatuses A, B, C, and D according to the fourth embodiment of the present invention are also characterized in that a printer that controls the split number of the Read request according to the latency and the plotter performance as described above is used. To do.

  FIG. 16 shows a data flow in the image forming system according to the fourth embodiment of the present invention. In FIG. 16, only the read data transfer from the print controller 1 is shown to simplify the drawing, the print instruction from the print controller 1 to the printing apparatuses A, B, C, and D, and the printing apparatuses A and B. , C, D from the flow of the read request to the print controller 1 is omitted.

  As shown in FIG. 16, the print controller 1 according to the present embodiment repeats a print job composed of a single page or a plurality of pages a plurality of times and prints multiple copies. Print control is executed in parallel for a plurality of printing apparatuses for the same print job as the input print job.

  A specific example will be described with reference to the flowchart of FIG. These operations are processed by software in the CPU 6 of the print controller 1 or implemented by hardware logic that performs equivalent operations. The operation from when a print job is input to the print controller 1 via the network 12 until print data of all pages is recorded in the HDD 8 is as described above.

  First, when print data for all pages is recorded in the HDD 8 (Yes in steps S21 to S25), the print controller 1 prints all the printing apparatuses A, B, C, and D connected to the PCIe switch 3. All the pages of the job are transferred in the order of 1, 2, 3, 4... Page from the first page (or the reverse order from the last page) (step S26).

  When the printing apparatuses A, B, C, and D start receiving print data, they sequentially execute print processing from the received page. Every time printing of the last page of the print job is completed, The number of copies that have been printed is notified from the endpoint 5 of the PCIe to the print controller 1 via the PCIe switch 3.

  The print controller 1 detects the number of print copies notified from each of the printing apparatuses A, B, C, and D while transferring the print data to each of the printing apparatuses A, B, C, and D (step S27). When the total number of copies notified from the printing apparatuses A, B, C, and D reaches the number of copies designated by the print job (Yes in step S28), the entire transfer process is terminated (step S29).

  As described above, according to the present embodiment, when the input print job is to print a plurality of copies by repeating a print job composed of a single page or a plurality of pages a plurality of times, the input print job Because the same print job is printed on multiple printing devices in parallel, even if the performance of each printing device is different, all printing devices do not always stop and complete printing at the maximum speed can do.

[Fifth Embodiment]
Next, a fifth embodiment of the present invention will be described with reference to FIGS. The same parts as those in the first to fourth embodiments described above are denoted by the same reference numerals, and description thereof is also omitted. That is, the printing apparatuses A, B, C, and D according to the fifth embodiment of the present invention are also characterized in that a printer that controls the number of splits of a Read request according to latency and plotter performance is used as described above. To do.

  First, in the aspect shown in the third embodiment, a case where a new print job is input while printing processing is being executed by a plurality of printing apparatuses will be described.

  FIG. 18 shows a data flow in the image forming system according to the fifth embodiment of the present invention. In FIG. 18, only the read data transfer from the print controller 1 is shown to simplify the drawing, the print instruction from the print controller 1 to the printing apparatuses A, B, C, and D, and the printing apparatuses A and B. , C, D from the flow of the read request to the print controller 1 is omitted.

  As shown in FIG. 18, the print controller 1 according to the present embodiment submits a new print job while printing processing is being executed by a plurality of printing apparatuses in the mode shown in the third embodiment. If the priority information specified for the print job is detected, and the priority is higher than the job being printed, the job of some printing devices is stopped and the newly submitted print The printing apparatus that has executed the job and did not stop printing continuously executes the print job that was being executed first.

  A specific example will be described with reference to the flowchart of FIG. These operations are processed by software in the CPU 6 of the print controller 1 or implemented by hardware logic that performs equivalent operations. The operation from when a print job is input to the print controller 1 via the network 12 until print data of all pages is recorded in the HDD 8 is as described above.

  First, when print data of all pages is recorded in the HDD 8 (Yes in steps S1 to S5), the print controller 1 changes the printing devices A, B, C, and D connected to the PCIe switch 3 to the printing devices A, A first group composed of B and a second group composed of printing apparatuses C and D are grouped (step S6).

  Next, the print controller 1 determines the page range to be printed by each group based on the performance of each printing apparatus (step S7). In the example shown in FIG. 18, assuming that the performances of the printing apparatuses A, B, C, and D are all the same, 10 pages of 20 pages of jobs are equally allocated to the first group and the second group. (That is, 1 to 10 pages are allocated to the first group, and 11 to 20 pages are allocated to the second group). If the performance of each printing apparatus is not uniform, it is desirable to assign a grouping and a print page range according to the speed. As an example, two printers are grouped in descending order of performance, and a wide print page range is assigned to high performance groups according to the total performance ratio of the printing devices in each group, while narrow printing is performed for low performance groups. By assigning the page range, the time for completing the printing of the entire print job can be shortened.

  After the printing device grouping and page range assignment are completed, the print controller 1 executes a process of transferring print data from the HDD 8 to each of the printing devices A, B, C, and D (step S8).

  Here, print data is transferred in ascending order from the first page of the assigned print page range to the first printing device in each group, and the first printing device is transferred to the second printing device. On the contrary, control is executed so that print data is transferred in reverse order from the last page in the assigned print page range. In the example shown in FIG. 18, print data is transferred to the first group of printing apparatuses A in the order 1, 2, 3,..., And the first group of printing apparatuses B is 10, 9, 8,. The print data is transferred in this order. On the other hand, the print data is transferred to the second group printing apparatus C in the order of 11, 12, 13,..., And the print data is transferred to the second group printing apparatus D in the order of 20, 19, 18. Will be transferred.

  When the printing apparatuses A, B, C, and D start receiving print data, the printing apparatuses sequentially execute print processing from the received page. However, each time the print data of one page is received, the page that has been printed. The page number is notified from the PCIe Endpoint 5 to the print controller 1 via the PCIe switch 3.

  The print controller 1 receives the page number notified from each printing device A, B, C, D while transferring the print data to each printing device A, B, C, D (step S9).

  Here, when a print job with a higher priority (priority job) is input later during execution of the print job transfer process, the print controller 1 detects this (Yes in step S31) and either The transfer processing for the printing apparatuses in the group is interrupted, and the print data for the priority job is switched to transfer (step S32). The example illustrated in FIG. 18 illustrates a state in which the second group of printing apparatuses D is switched to transfer of print data of a priority job. Since the priority job starts printing immediately by the printing apparatus D as it is, the print job can be completed without waiting until all the previously input print jobs are completed. At this time, since the printing device C in the second loop continues to print the print data of the print job that has been input as it is, the assigned printing is performed even when the printing device D is interrupted by the priority job. It becomes possible to print the print data of the entire page range.

  After that, if there is a group in which a state in which the first and second page numbers match on the next page is detected (Yes in step S10), the transmission of print data to the group is completed (step S11). . In each group, two printers are printing from the first page and the last page respectively, so if the above condition is detected, all pages assigned to that group have been printed. become. With the above operation, when the printing process is advanced and the printing process in all groups is completed (Yes in step S12), that is, the printing of all pages of the input print job is completed, and the entire process is finished. (Step S13).

  Next, in the aspect shown in the fourth embodiment, a case where a new print job is input while printing processing is being executed by a plurality of printing apparatuses will be described.

  FIG. 20 shows a data flow in the image forming system according to the fifth embodiment of the present invention. In FIG. 20, only the read data transfer from the print controller 1 is shown to simplify the drawing, the print instruction from the print controller 1 to the printing apparatuses A, B, C, and D, and the printing apparatuses A and B. , C, D from the flow of the read request to the print controller 1 is omitted.

  As shown in FIG. 20, the print controller 1 according to the present embodiment submits a new print job while printing processing is being executed by a plurality of printing apparatuses in the mode shown in the fourth embodiment. If the priority information specified for the print job is detected, and the priority is higher than the job being printed, the job of some printing devices is stopped and the newly submitted print The printing apparatus that has executed the job and did not stop printing continuously executes the print job that was being executed first.

  A specific example will be described with reference to the flowchart of FIG. These operations are processed by software in the CPU 6 of the print controller 1 or implemented by hardware logic that performs equivalent operations. The operation from when a print job is input to the print controller 1 via the network 12 until print data of all pages is recorded in the HDD 8 is as described above.

  First, when print data for all pages is recorded in the HDD 8 (Yes in steps S21 to S25), the print controller 1 prints all the printing apparatuses A, B, C, and D connected to the PCIe switch 3. All the pages of the job are transferred in the order of 1, 2, 3, 4... Page from the first page (or the reverse order from the last page) (step S26).

  When the printing apparatuses A, B, C, and D start receiving print data, they sequentially execute print processing from the received page. Every time printing of the last page of the print job is completed, The number of copies that have been printed is notified from the endpoint 5 of the PCIe to the print controller 1 via the PCIe switch 3.

  The print controller 1 detects the number of print copies notified from each of the printing apparatuses A, B, C, and D while transferring the print data to each of the printing apparatuses A, B, C, and D (step S27). When the total number of copies notified from the printing apparatuses A, B, C, and D reaches the number of copies designated by the print job (Yes in step S28), the entire transfer process is terminated (step S29).

  If a higher priority print job (priority job) is input later during execution of the print job transfer process, the print controller 1 detects this (Yes in step S41) and either The transfer process for the printing apparatuses in the group is interrupted, and the print data of the priority job is switched to transfer (step S42). The example shown in FIG. 20 shows a state in which the printing device D of the second group is switched to the transfer of the print data of the priority job. However, the printing devices A, B, and C continue to print previously input. Since job printing continues to be executed, printing of all copies can be completed in parallel with the priority job.

  As described above, according to the present embodiment, even when a job that is to be prioritized over a job being printed is submitted later, the jobs of some printing apparatuses are stopped and the newly submitted print job is Since the printing apparatus that has been executed and has not stopped printing continuously executes the print job that was being executed first, both print jobs can be processed at high speed.

[Sixth Embodiment]
Next, a sixth embodiment of the present invention will be described with reference to FIGS. The same parts as those in the first to fifth embodiments described above are denoted by the same reference numerals, and description thereof is also omitted. That is, the printing apparatuses A, B, C, and D according to the sixth embodiment of the present invention are also characterized in that a printer that controls the number of Read requests according to the latency and plotter performance is used as described above. To do.

  First, in the aspect shown in the third embodiment, during the execution of printing processing in a plurality of printing apparatuses, paper formation, that is, consumables (toner, ink, paper) run out, and other problems such as failure of image formation. A case where a possible failure occurs will be described.

  FIG. 22 shows a data flow in the image forming system according to the sixth embodiment of the present invention. In FIG. 22, only the read data transfer from the print controller 1 is shown to simplify the drawing, the print instruction from the print controller 1 to the printing apparatuses A, B, C, and D, and the printing apparatuses A and B. , C, D from the flow of the read request to the print controller 1 is omitted.

  As shown in FIG. 22, the print controller 1 according to the present embodiment, in the aspect shown in the third embodiment, is a paper, that is, a consumable ( When a failure that prevents image formation due to toner, ink, paper) or other failure occurs, the job of the printing device that detected this is stopped, and the print job that was being executed on the printing device that did not stop printing Is executed continuously.

  A specific example will be described with reference to the flowchart of FIG. These operations are processed by software in the CPU 6 of the print controller 1 or implemented by hardware logic that performs equivalent operations. The operation from when a print job is input to the print controller 1 via the network 12 until print data of all pages is recorded in the HDD 8 is as described above.

  First, when print data of all pages is recorded in the HDD 8 (Yes in steps S1 to S5), the print controller 1 changes the printing devices A, B, C, and D connected to the PCIe switch 3 to the printing devices A, A first group composed of B and a second group composed of printing apparatuses C and D are grouped (step S6).

  Next, the print controller 1 determines the page range to be printed by each group based on the performance of each printing apparatus (step S7). In the example shown in FIG. 22, assuming that the performances of the printing apparatuses A, B, C, and D are all the same, 10 pages of 20 pages of jobs are equally allocated to the first group and the second group. (That is, 1 to 10 pages are allocated to the first group, and 11 to 20 pages are allocated to the second group). If the performance of each printing apparatus is not uniform, it is desirable to assign a grouping and a print page range according to the speed. As an example, two printers are grouped in descending order of performance, and a wide print page range is assigned to high performance groups according to the total performance ratio of the printing devices in each group, while narrow printing is performed for low performance groups. By assigning the page range, the time for completing the printing of the entire print job can be shortened.

  After the printing device grouping and page range assignment are completed, the print controller 1 executes a process of transferring print data from the HDD 8 to each of the printing devices A, B, C, and D (step S8).

  Here, print data is transferred in ascending order from the first page of the assigned print page range to the first printing device in each group, and the first printing device is transferred to the second printing device. On the contrary, control is executed so that print data is transferred in reverse order from the last page in the assigned print page range. In the example shown in FIG. 22, print data is transferred to the first group of printing apparatuses A in the order 1, 2, 3,..., And the first group of printing apparatuses B is 10, 9, 8,. The print data is transferred in this order. On the other hand, the print data is transferred to the second group printing apparatus C in the order of 11, 12, 13,..., And the print data is transferred to the second group printing apparatus D in the order of 20, 19, 18. Will be transferred.

  When the printing apparatuses A, B, C, and D start receiving print data, the printing apparatuses sequentially execute print processing from the received page. However, each time the print data of one page is received, the page that has been printed. The page number is notified from the PCIe Endpoint 5 to the print controller 1 via the PCIe switch 3.

  The print controller 1 receives the page number notified from each printing device A, B, C, D while transferring the print data to each printing device A, B, C, D (step S9).

  Here, during execution of the print job transfer process, a failure such as a failure or a consumable failure may occur in any of the printing apparatuses. In the example illustrated in FIG. 22, it is assumed that a failure has occurred in the second group of printing apparatuses D. When a failure / consumable item runs out during printing of a print job, the printing apparatus D stops the printing process and notifies the print controller 1 from the PCIe Endpoint 5 via the PCIe switch 3. For other printing apparatuses as well, when a failure occurs, the print controller 1 is similarly notified of failure / consumables.

  If a failure / out of consumables notification is received during execution of the print job transfer process, the print controller 1 detects this (Yes in step S51) and stops the print data transfer process for the notification source printing apparatus. (Step S52). In the example shown in FIG. 22, the transfer of print data to the second group of printing apparatuses D is interrupted. At this time, since the printing device C of the second group continues to print the print data of the print job that has been input as it is, the assigned print page range even when the printing device D is interrupted during printing. All print data can be printed as it is.

  After that, if there is a group in which a state in which the first and second page numbers match on the next page is detected (Yes in step S10), the transmission of print data to the group is completed (step S11). . In each group, two printers are printing from the first page and the last page respectively, so if the above condition is detected, all pages assigned to that group have been printed. become. With the above operation, when the printing process is advanced and the printing process in all groups is completed (Yes in step S12), that is, the printing of all pages of the input print job is completed, and the entire process is finished. (Step S13).

  Next, in the aspect shown in the fourth embodiment, while printing processing is being executed in a plurality of printing apparatuses, paper, that is, consumables (toner, ink, paper) run out, or other troubles due to failure occur. The case where it occurs will be described.

  FIG. 24 shows a data flow in the image forming system according to the sixth embodiment of the present invention. In FIG. 24, only the read data transfer from the print controller 1 is shown to simplify the drawing, the print instruction from the print controller 1 to the printing apparatuses A, B, C, and D and the printing apparatuses A and B , C, D from the flow of the read request to the print controller 1 is omitted.

  As shown in FIG. 24, in the aspect shown in the fourth embodiment, the print controller 1 according to the present embodiment is paper, that is, a consumable ( When a failure that prevents image formation due to toner, ink, paper) or other failure occurs, the job of the printing device that detected this is stopped, and the print job that was being executed on the printing device that did not stop printing Is executed continuously.

  A specific example will be described with reference to the flowchart of FIG. These operations are processed by software in the CPU 6 of the print controller 1 or implemented by hardware logic that performs equivalent operations. The operation from when a print job is input to the print controller 1 via the network 12 until print data of all pages is recorded in the HDD 8 is as described above.

  First, when print data for all pages is recorded in the HDD 8 (Yes in steps S21 to S25), the print controller 1 prints all the printing apparatuses A, B, C, and D connected to the PCIe switch 3. All the pages of the job are transferred in the order of 1, 2, 3, 4... Page from the first page (or the reverse order from the last page) (step S26).

  When the printing apparatuses A, B, C, and D start receiving print data, they sequentially execute print processing from the received page. Every time printing of the last page of the print job is completed, The number of copies that have been printed is notified from the endpoint 5 of the PCIe to the print controller 1 via the PCIe switch 3.

  The print controller 1 detects the number of print copies notified from each of the printing apparatuses A, B, C, and D while transferring the print data to each of the printing apparatuses A, B, C, and D (step S27). When the total number of copies notified from the printing apparatuses A, B, C, and D reaches the number of copies designated by the print job (Yes in step S28), the entire transfer process is terminated (step S29).

  If a failure / consumable out-of-stock notification is received during print job transfer processing, the print controller 1 detects this (Yes in step S61) and performs print data transfer processing for the notification source printing apparatus. Stop (step S62). In the example shown in FIG. 24, the transfer of print data to the second group of printing apparatuses D is interrupted. At this time, since the printing device C of the second group continues to print the print data of the print job that has been input as it is, the assigned print page range even when the printing device D is interrupted during printing. All print data can be printed as it is.

  As described above, according to the present embodiment, while printing processing is being performed in the printing apparatus, paper, that is, a consumable (toner, ink, paper) is out of print, or an unprintable failure occurs due to other failures. In this case, the job of the printing apparatus that detected the failure is stopped, and the print job that was being executed by the printing apparatus that did not stop printing is continuously executed. Therefore, both print jobs can be processed at high speed.

1 is a block diagram showing a configuration of an image forming system according to a first embodiment of the present invention. FIG. 4 is a schematic diagram illustrating a data flow until image data for printing is stored in an HDD. It is a schematic diagram which shows the flow of the data at the time of printing. It is a timing chart which shows the operation example in case the number of splits is "1". 10 is a timing chart showing an operation example when the number of splits is “2”. 10 is a timing chart showing an operation example when the number of splits is “2”. 12 is a timing chart showing an example of operation when the number of splits is “3”. It is a graph which shows the relationship between the latency concerning the 2nd Embodiment of this invention, and the number of splits. 6 is a graph showing the relationship between the maximum data transfer bandwidth of a plotter and the number of splits. It is a graph which shows the relationship between the number of splits, latency, and plotter bandwidth. It is a graph which shows the relationship between the number of plotter starting, and latency. It is a schematic diagram which shows the relationship between the number of plotter starting, and latency. It is a block diagram which shows the modification of a structure of an image forming system. It is a schematic diagram which shows the data flow in the image forming system concerning the 3rd Embodiment of this invention. 6 is a flowchart illustrating a flow of printing processing. It is a schematic diagram which shows the data flow in the image forming system concerning the 4th Embodiment of this invention. 6 is a flowchart illustrating a flow of printing processing. It is a schematic diagram which shows the data flow in the image forming system concerning the 5th Embodiment of this invention. 6 is a flowchart illustrating a flow of printing processing. It is a schematic diagram which shows the flow of the data in an image forming system. 6 is a flowchart illustrating a flow of printing processing. It is a schematic diagram which shows the data flow in the image forming system concerning the 6th Embodiment of this invention. 6 is a flowchart illustrating a flow of printing processing. It is a schematic diagram which shows the flow of the data in an image forming system. 6 is a flowchart illustrating a flow of printing processing.

Explanation of symbols

1 Control means 6 CPU
10 memory 12 network 21 image forming means 22 DMAC

Claims (9)

A memory for storing image data;
A single memory or a plurality of DMACs (Direct Memory Access Controllers) that perform data transfer to and from the memory without using a CPU that controls the entire system, and that perform image formation of the image data stored in the memory Image forming means;
The image data that is connected to the image forming unit via a high-speed serial transmission path and stores the image data relating to a print job submitted via a network is stored in the memory and requested by a request issued by the DMAC. Control means for transferring data from the memory to the requesting image forming means;
With
The image forming unit operates the split number of requests, which is the number of subsequent requests issued continuously while the image data for a predetermined request returns from the control unit. Change according to the latency of the request, which is the delay time until receiving the image data for the request in the state,
An image forming system. The image forming unit detects a latency of a read request when requesting transfer of the image data to the control unit by a memory read request, and determines a split number of the request as the detected latency is larger. A split issuance control means that increases and decreases the number of split requests as the detected latency is smaller,
The image forming system according to claim 1. The image forming unit includes: a latency detecting unit that detects a latency of a read request when requesting the control unit to transfer the image data by a memory read request; and a maximum data transfer for image formation of the image forming unit. A split issuance control means for increasing the number of splits as the bandwidth increases, and decreasing the number of splits as the maximum data transfer bandwidth decreases;
The image forming system according to claim 1. The control unit divides the input print job into a plurality of the image forming units and executes control, and when the input print job is composed of a plurality of pages, one print job is Dividing the plurality of image forming units to control execution of image formation;
The image forming system according to claim 1, wherein the image forming system is an image forming system. The control unit divides a plurality of image forming units into a plurality of groups each by dividing a plurality of the image forming units when performing control by dividing a print job composed of a plurality of pages with respect to the plurality of image forming units, A range composed of a plurality of pages is assigned to each group, and the first image forming means of each group controls execution of image formation in order from the first page of the assigned range, and the second unit The image forming means controls execution of image formation in reverse order from the last page of the allocated range;
The image forming system according to claim 4. The control means is the same as the input print job when the input print job is to execute image formation of multiple copies by repeating a print job composed of a single page or a plurality of pages a plurality of times. Executing image forming control on a plurality of image forming units in parallel with a print job;
The image forming system according to claim 4 or 5, wherein The control unit detects priority information specified in the print job and performs image formation when a new print job is input while performing image forming processing in the plurality of image forming units. If the priority is higher than the middle job, a part of the job of the image forming unit is stopped, a newly input print job is executed, and the image forming unit that has not stopped printing is first. Continue the print job that was being executed,
The image forming system according to claim 4, wherein the image forming system is an image forming system. The control unit stops a job of the image forming unit that detects the occurrence of a failure that prevents image formation when an image formation process is being performed in the plurality of image forming units, and image formation is performed. The image forming unit that has not stopped processing continuously executes the print job that was being executed.
The image forming system according to claim 4, wherein the image forming system is an image forming system. Computer
A DMAC (Direct Memory Access Controller) that transfers data to and from a memory that stores image data without using a CPU that controls the entire system, and that performs image formation of the image data stored in the memory. The image data relating to a print job that is connected to one or a plurality of image forming means via a high-speed serial transmission line and is input via the network is stored in the memory and requested by a request issued by the DMAC. A program for causing the image data to function as a control unit that transfers the image data from the memory to the requesting image forming unit,
The image forming unit operates the split number of requests, which is the number of subsequent requests issued continuously while the image data for a predetermined request returns from the control unit. Change according to the latency of the request, which is the delay time until receiving the image data for the request in the state,
A program characterized by that.

Images