JP2012181578A - Update control device and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To efficiently perform update by suppressing load concentration on hardware resources (hereinafter, resources) when updating software (hereinafter, SW) of a plurality of operating systems (hereinafter, OS).SOLUTION: Based on each piece of attribute information of pieces of update SW that update each piece of SW of a plurality of OSs operating by sharing resources and each piece of information about the resources allocated to each of the plurality of OSs, an update control device estimates time required for each piece of processing to be executed when each piece of update SW performs update. Based on the estimation, the device determines execution order and execution start timing for each piece of processing to be executed at update so as to reduce time for executing pieces of processing having the same processing contents in parallel in the plurality of OSs. According to the estimated time, determined execution order and execution start timing, the device determines an allocation rate of the resources for each of the plurality of OSs. Then, the device controls each piece of processing of the update SW to be executed by determined execution order and at the determined execution start timing, and controls the resources to be allocated by the determined allocation rates.

Description

  The present invention relates to an update control device and a program.

  Patent Document 1 discloses an access control method for controlling access to a memory block shared among a plurality of processes when a plurality of processes are repeatedly executed for each process at a predetermined cycle. Accepts the setting of access time to the block, sets an access period to the memory block for each process based on a predetermined cycle and the set access time, and controls access to the memory block according to the set access period An access control method is disclosed.

  In Patent Document 2, in a computer resource allocation method of a computer system in which computer resources are allocated to a plurality of computers and a program is executed independently on each computer, the resource usage status of the computers is collected and the collected data Based on the collected data and the calculated correlation coefficient, the resource allocation value of each computer is calculated, and each computer is calculated according to the resource allocation value. A computer resource allocation method for performing resource allocation is disclosed.

JP 2000-148777 A JP 2004-199561 A

  The present invention parallelizes software updates of a plurality of operating systems without controlling the execution order, execution start timing, and hardware resource allocation of each process executed when updating the software of a plurality of operating systems. It is an object of the present invention to obtain an update control device and a program that can suppress the load concentration of hardware resources and efficiently perform update processing as compared to the case where the update processing is performed.

  The update control apparatus according to claim 1 is an update control apparatus provided in an information processing apparatus in which a plurality of operating systems operate by sharing hardware resources, and each of the software of the plurality of operating systems. Is executed at the time of update by each update software based on each of the attribute information of the update software for updating the software and each of the hardware resource information assigned to each of the plurality of operating systems A prediction means for predicting the time required for each process; and a case where the plurality of operating systems start execution of the update software at the same time and update the software of each operating system based on the predicted time. Compared to multiple operating systems First determination means for determining the execution order and execution start timing of each of the processes executed when updating by the update software so that the time during which the processes having the same content are executed in parallel is shortened, and the predicted Second determination means for determining an allocation rate of the hardware resources for each of the plurality of operating systems according to time, the execution order, and the execution start timing; and the determined execution order and execution start timing And a control unit that performs control so that each process by the update software is executed and the hardware resources are allocated at the determined allocation rate.

  According to a second aspect of the present invention, in the update control apparatus according to the first aspect, the control means performs prediction by the prediction means at predetermined time intervals even during execution of each process by the update software. The execution order and execution start timing determined by the first determining means and the allocation rate determined by the second determining means are controlled, and the execution order and execution start timing newly determined by the control are controlled. In this case, each process by the update software is executed, and control is performed so that the hardware resource is allocated at an allocation rate newly determined by the control.

  According to a third aspect of the present invention, update software attribute information for updating each of software of the plurality of operating systems is provided on a computer provided in an information processing apparatus in which a plurality of operating systems operate by sharing hardware resources. And the time required for each of the processes executed when updating by each update software is predicted based on each of the plurality of operating systems and information of hardware resources allocated to each of the plurality of operating systems The plurality of operating systems based on the predicted time, as compared with a case where the plurality of operating systems start execution of the updated software all at once and update software of each operating system based on the predicted time. Processes with the same processing content between systems are parallel First determination means for determining the execution order and execution start timing of each of the processes executed during the update by the update software so that the execution time is shortened, the predicted time, the execution order, and the execution A second determination unit configured to determine an allocation rate of the hardware resource for each of the plurality of operating systems according to a start timing; and each of the processes is executed in the determined execution order and execution start timing. This is a program for functioning as control means for controlling the hardware resources to be allocated at the allocated rate.

  According to the first aspect of the present invention, a plurality of operating systems can be controlled without controlling the execution order, execution start timing, and hardware resource allocation of each process executed when updating software of a plurality of operating systems. Compared with the case where the system software is updated in parallel, the load concentration of hardware resources can be suppressed and the update process can be performed efficiently.

  According to the second aspect of the present invention, the load concentration of hardware resources can be suppressed and the update process can be performed efficiently compared with the case where this configuration is not provided.

  According to the third aspect of the present invention, a plurality of operating systems can be controlled without controlling the execution order, execution start timing, and hardware resource allocation of each process executed when updating software of a plurality of operating systems. Compared with the case where the system software is updated in parallel, the load concentration of hardware resources can be suppressed and the update process can be performed efficiently.

1 is a diagram illustrating a hardware configuration of an image forming apparatus according to an embodiment. 2 is a diagram illustrating a software configuration of the image forming apparatus according to the embodiment. FIG. It is explanatory drawing explaining the function of a hypervisor. It is a flowchart which shows the flow of the processing routine performed with a hypervisor. (A) shows an example of attribute information of an update package, and (B) is a diagram showing a specific example of performance information of a virtual machine. It is a flowchart which shows the flow of the scheduling process performed with a hypervisor. It is explanatory drawing explaining the determination of the update schedule and the allocation rate of a hardware resource. It is the figure which showed a mode that load concentrated. It is explanatory drawing explaining an example of the determination method of execution start timing. It is explanatory drawing explaining an example of the determination method of an execution order and execution start timing. It is a figure which shows a host OS type software structure.

  Hereinafter, an example of an embodiment will be described in detail with reference to the drawings. FIG. 1 shows an image forming apparatus 10 according to the present exemplary embodiment. The image forming apparatus 10 includes an apparatus control unit 12 that controls the operation of each unit of the image forming apparatus 10, and an image reading unit that optically reads a set document (paper document) to be read and outputs read image data. 14, an image forming unit 16 that forms an image represented by input image data on a recording sheet, a display unit 18 A composed of an LCD or the like, and an operation accepting unit 18 B that includes a numeric keypad, a touch panel, etc. and accepts an operation by a user. And an information processing apparatus such as a PC (Personal Computer) and a network interface 20 for transmitting and receiving information via a network cable and communication means (for example, a computer network). Are connected to each other via a bus 22.

  The device control unit 12 includes a microcomputer and the like, and is provided with a non-volatile storage unit 12C including a CPU 12A, a memory 12B, an HDD (Hard Disk Drive), a flash memory, and the like. In the present embodiment, this storage unit 12C is an HDD. The CPU 12A may be a single core or a multi-core. The memory 12B is used as a work area, for example. The storage unit 12 </ b> C shows application programs (a first application program and a second application program in FIG. 1) that perform processing for providing the functions of the image forming apparatus 10 to the user of the image forming apparatus 10. ) And an operating system (OS) program that functions as a platform for executing the application program (the first OS program and the second OS program are shown in FIG. 1). In the present embodiment, a plurality of OS programs are stored in the storage unit 12C, and each application is installed on any one OS and operates on the OS (see also FIG. 2).

  As shown in FIG. 1, the storage unit 12C also stores a hypervisor program that provides functions such as hardware resource virtualization and shared resource arbitration with a plurality of OSs. Then, as shown in FIG. 2, the first virtual machine is constructed so that the first application operates on the first OS, and the second virtual machine is constructed so that the second application operates on the second OS.

  As shown in FIG. 3, when assigning the CPU 12A to a plurality of OSs, the hypervisor defines a virtual CPU for each virtual machine and divides the processing time of the real CPU (CPU 12A) into short units. The virtual CPUs are allocated in a time division manner. It is also possible to increase or decrease the virtual CPU allocation rate while operating the OS. When the CPU 12A is multi-core, the CPU 12A may be assigned to each OS by adjusting the number of cores to be assigned.

  As for the assignment of the network interface 20, as shown in FIG. 3, a virtual Net is defined for each virtual machine. The virtual Net is connected to the real Net (network interface 20) via a virtual switch or the like, for example, and uses the real Net. The plurality of virtual Nets share the network bandwidth (data transfer amount per unit time) of the real Net allocated to each of them.

  As for the assignment of the HDD 22, as shown in FIG. 3, a virtual HDD is defined for each virtual machine. Each virtual HDD is allocated to a real HDD (storage unit 12C), and shares the disk I / O bandwidth (read / write data amount per unit time) of the real HDD.

  Note that each of the hypervisor and the plurality of OSs communicate with each other using a known inter-processor communication technique. For example, information may be transferred by copying a data packet from a transmission source (transmission side) to a reception buffer of a transmission destination (reception side), or a shared memory shared by the hypervisor and each OS. Information may be transferred via (a predetermined storage area of the memory 12B).

  By the way, each OS installed in an information processing apparatus such as the image forming apparatus 10 may require an OS software update as shown in FIG.

  In the case of a system in which a plurality of virtual machines are installed in an embedded device such as a home appliance or an office device as in this embodiment, the OS software is updated in order to match each function and reduce the frequency of stopping each function. In such a case, it is often necessary to update not only a part of the OS but all the OSs. When updating a plurality of OSs, update each OS one by one (the OSs are updated individually so that the software update periods do not overlap at all), and the first OS starts updating. Since the time required for the last OS to complete the update becomes longer, it is more efficient that a plurality of OSs operate in parallel to execute the update. Therefore, in this embodiment, it is assumed that a plurality of OSs update software in parallel.

  Hereinafter, the operation at the time of software update of the OS according to the present embodiment will be described. Hereinafter, a collection of update software files for updating the OS is referred to as an update package. Note that the number of update packages used for the update for each OS is not limited to one and may be plural. Further, during the update execution, various processes such as a process of downloading the update package from the server, a process of updating the software with the downloaded update package (that is, a process of replacing the old software with the new software), a process of initializing the database, etc. In this embodiment, a series of processes from the start of downloading an update package to the end of the last process by the downloaded update package is referred to as update or update process. A process for replacing old software with new software is referred to as an update process in order to distinguish it from the update process.

  FIG. 4 is a flowchart showing the flow of a processing routine executed by the hypervisor. In step 100, attribute information of the update package of each OS is acquired from a server or the like that holds the update package of each OS via the network. The file indicating the attribute information is included in the update package in advance, and the attribute information file of the update package is transmitted from the server in response to an inquiry transmitted from the hypervisor to the server.

  Here, the attribute information includes, for example, the package size, the presence / absence of dependent packages, the processing and processing sequence executed during the update, the time required for each processing at the time of updating (required for each processing when the update is executed on the reference machine) Time), network access amount at the time of update (network communication amount when update is executed on the reference machine), performance value of the reference machine, and the like.

  The package size can be the total value of the file sizes of software included in the update package.

  In addition, whether or not there is a dependency package, for example, when it is necessary to update using two update packages, after the update process by one update package is completed, the update process by the other update package is executed Is determined, the one update package becomes a “dependency package” for the other update package. The same applies when updating using three or more update packages.

  The process executed during the update is, for example, as described above, a download process, an update process, a restart process (the updated software may be restarted or the OS may be restarted, Hereinafter, the former is referred to as SW restart, the latter is referred to as SY restart, and database initialization processing.

  FIG. 5A shows an example of attribute information of the update package A. In FIG. 5A, the size is the package size, the update processing content is the processing and processing order executed during the update, and the restart waiting time is the restart executed during the update. It takes time.

  In FIG. 5A, the CPU performance reference value, the memory access speed reference value, the disk access speed reference value, and the network speed reference value are the CPU performance reference value, memory access speed reference value, and disk access speed of the reference machine. Reference value and network speed reference value. The restart waiting time and update network access amount in the table are the time and access amount when executed on the reference machine indicated by these values. Here, the disk refers to the HDD.

  Next, in step 102, the current performance information of each virtual machine is acquired. Here, the current performance information of the virtual machine is information on hardware resources allocated to each virtual machine (information on hardware resources allocated to each OS constituting each virtual machine. In other words). Although the hardware resource allocation state dynamically changes, here, performance information in the allocation state at the time of execution of step 102 is acquired.

  FIG. 5B shows a specific example of the performance information of the virtual machine. Here, each of the CPU performance, memory access performance, disk access performance, and network speed is a performance value of the hardware resource itself that is shared and used by each virtual machine, and is a value common to each virtual machine. . However, since the CPU allocation rate, disk I / O bandwidth, and network bandwidth have different allocation states for each virtual machine, these values are values for each virtual machine.

  In step 104, scheduling processing is performed. FIG. 6 shows a detailed flowchart of the scheduling process executed by the hypervisor. Here, it is assumed that the image forming apparatus 10 is provided with two virtual machines, the first OS operates on one virtual machine, and the second OS operates on the other virtual machine.

  In step 200, it is determined whether there is an update package for the first OS. If it is determined in step 200 that the first OS has an update package, one update package for the first OS is selected in step 202. Selection is made based on attribute information of the update package. For example, in the example shown in FIG. 5A, “package A is executed after package B of version 2.0 or later” as shown in the item “dependency package”. Therefore, at least the update package B is selected as an update target before the update package A and is selected to be processed. In the present embodiment, the schedule is determined so that the update processing by each update package is performed in the order of selection of the update packages.

  In step 204, the time (predicted time) required for each process performed in the update by the selected update package is predicted based on the acquired attribute information of the update package and the performance information of the virtual machine. When the attribute information of the update package is the attribute information shown in FIG. 5A and the performance information of the virtual machine is the performance information shown in FIG. 5B, for example, it is calculated by the prediction formula shown below.

・ Predicted download time = k1 x Sa x (P1 / Pa1 x R1) x (P2 / Pa2) x (P3 / Pa3 x R3)
-Estimated update time = k2 x Sa x (P1 / Pa1 x R1) x (P2 / Pa2) x (P3 / Pa3 x R3) + k3 x Na x (P4 / Pa4 x R4)
-SW restart estimated time = Ta x (P1 / Pa1 x R1) x (P2 / Pa2) x (P3 / Pa3 x R3)
Here, each of k1, k2, and k3 is a predetermined coefficient.

  Note that the above is an example of a simple prediction formula, and the download prediction time and update prediction time are the package size (approximately equal to the sum of the software file sizes to be newly replaced) and the hardware performance ratio of the reference machine and the virtual machine. It is a formula calculated from Further, the predicted SW restart time is an expression obtained by multiplying the restart waiting time included in the attribute information by the hardware performance ratio.

  Actually, there is a correlation between the items of the performance information of the virtual machine, but the prediction formula is not a formula that takes the correlation into account. Although illustration is omitted, the prediction time of each process may be calculated using a prediction formula in consideration of the correlation. Thereby, a more accurate prediction time is calculated.

  Further, the CPU allocation rate, the disk I / O bandwidth, and the network bandwidth of the virtual machine that are used in the prediction formula need to be defined for each process. In the prediction formula illustrated above, execution of step 102 is performed. The performance information value acquired at the time is used. However, the present invention is not limited to this, and may be a value determined in advance according to the processing content. Also, for example, when the download process and the update process are executed in parallel between the first OS and the second OS (see also FIG. 9), the CPU allocation rate is set to the first OS update process: the second OS download process = It is set in advance such as 30:70. As a result, the CPU allocation rate, disk I / O bandwidth, and network bandwidth become different values for each process. In this case, since the schedule of the second OS has not yet been determined at the time of determining the allocation rate for the first OS, it is preferable to make a provisional determination and make adjustments later.

  After calculating the predicted time, the execution order and execution start timing (schedule) of each process are determined. The first OS is scheduled so that the update process of the next update package is started after the update process of one update package is completed. Therefore, in the present embodiment, the execution order is determined so that the update processing is performed in the order of selection of update packages in step 202 described above. In addition, the execution order of each process executed in each update package is determined to be executed in the order of processes defined in the update package attribute information.

  In step 206, the allocation rate of hardware resources for the first OS is determined for each period in which each process of the update package is executed. For example, a predetermined value may be determined as the allocation rate according to the processing content. More specifically, in the case of download processing, an allocation rate is determined and determined in advance, such as a network bandwidth of 70% and a CPU allocation rate of 30%.

  Further, the allocation rate may be determined in consideration of the schedule of each process between the first OS and the second OS. For example, as described above, when it is predetermined that the download process with a relatively low CPU load and the update process with a relatively high CPU load are executed in parallel between the first OS and the second OS, the second OS The CPU allocation rate during the period during which the download process is performed and the update process is performed by the first OS is set such that the second OS download process: the first OS update process = 30: 70. In this case, the schedule on the second OS side is related to the determination of the allocation rate. However, before the schedule on the second OS side is determined, the allocation rate of the hardware resources on the first OS side is tentatively determined here. However, it may be adjusted after the schedule on the second OS side is determined.

  As described above, as shown in FIG. 7, the schedule and hardware resource allocation rate for the selected update package are determined.

  After step 206, the process returns to step 200 and the above procedure is repeated for the next update package. When the above procedure for all the update packages of the first OS is completed, a negative determination is made in step 200, and the process proceeds to step 210.

  In step 210, it is determined whether there is an update package for the second OS. If it is determined in step 210 that the first OS has an update package, one update package for the second OS is selected in step 212. The selection method is the same as in step 202.

  In step 214, as in step 204, the time (predicted time) required for each process performed in updating the selected update package is predicted based on the acquired update package attribute information and virtual machine performance information. After calculating the predicted time, a schedule (execution order and execution start timing of each process) is determined. The schedule of the second OS is determined with reference to the schedule of the first OS.

  For example, if a process with the same content is executed in the same time zone on the first OS and the second OS, the process with the same content is similar or equal in the usage tendency of hardware resources. The load is concentrated on the hardware resources. Therefore, here, compared to the case where the first OS and the second OS start updating all at once, the time required for executing the processes having the same processing contents in the first OS and the second OS in parallel is shortened. The execution start timing of each process on the side is determined.

  For example, as shown in FIG. 8, since the download process has a large network I / O load, if the download process is executed on the second OS while the download process is executed on the first OS, the network I / O load increases. As a result, the processing efficiency decreases. In addition, since the load on the CPU 12A and the HDD is large in the update process, if the update process is executed on the second OS while the update process is executed on the first OS, the CPU / HDD load increases and the processing efficiency decreases. Therefore, the execution start timing of each process executed on the second OS side is adjusted so that the time during which processes having the same process contents are executed in parallel between the first OS and the second OS is shortened. For example, the execution start timing of the second OS download process is adjusted after the first OS download process ends or when the first OS download process ends at a predetermined rate. .

  In this way, by controlling so that the time when the processes having the same contents are executed in parallel is shortened, the use tendency of hardware resources is more than that when the first OS and the second OS start updating all at once. The time during which similar or equal processes are executed in parallel is shortened.

  Further, a schedule may be created as “no processing” during a period when there is no selectable processing (processing to be executed).

  For example, the update by the update package cannot be performed after the process of downloading the update package from the server is not performed first. Therefore, both the first OS and the second OS need to perform the download process first. For example, when adjusting so that processes with the same contents between OSs do not overlap in the same time period (period), the first OS side The fact that the download process is not executed on the second OS side during the time period when the first download process is executed means that there is no process that can be executed on the second OS side during that time period. Therefore, as shown in FIG. 9, the second OS side schedules “no processing” during that period. Then, after the download process is completed on the first OS side, the execution start timing is determined so that the download process is executed on the second OS side.

  In addition, even if scheduling as described above cannot be performed within a predetermined time for processing with the same content in parallel, within a range that does not violate the “dependence package” of the attribute information. The execution order of the update packages may be changed (returning to the process of step 212 and selecting the update package again). Alternatively, the update package may be selected again from the first OS side and scheduling may be performed again.

  In step 216, as in step 206, the allocation rate of hardware resources for the first OS is determined for each period in which each process of the update package is executed. The determination of the hardware resource allocation rate for the second OS is also referred to the schedule of the first OS, and in some cases, the work of adjusting the hardware resource allocation rate already determined by the first OS is also performed. . For example, a large amount of CPU occupation time is allocated to an OS that performs an update process with a high CPU load such as restart or database construction. When there is a difference in the update package size, a large amount of CPU (time and number of cores) and memory are allocated to an OS with a large amount of update (large size). Further, depending on the schedule, the hardware resource allocation may be determined to be 100% for one OS. In this case, in actual allocation control (step 106 to be described later), it is not used as virtual hardware (virtual HDD, virtual network,...) Via the hypervisor driver, but directly without using the hypervisor. You may make it switch the usage method of hardware so that real hardware may be used. The latter method of use is faster in processing speed than the former method of use.

  In addition, when the period when the sum total of the allocation rate of 1st OS and 2nd OS exceeds 100% occurs, the allocation rate with respect to each OS of the period is reduced by the equal ratio in each OS, and allocation rate Is adjusted so that the total of 100% or less.

  After step 216, the process returns to step 210, and the above procedure is repeated for the next update package. When the above procedure for all the update packages of the second OS is completed, a negative determination is made in step 210, and this processing routine ends.

  When the number of OSs is three or more, the processing procedure of steps 210 to 216 of the second OS may be repeated for each of the third OS, the fourth OS,.

  In addition, the update package execution order is changed within a range that does not violate the “dependence package” of the attribute information so that the update end timings match between the OSs or the time difference is within a predetermined range (step 202 or above). It is also possible to return to the process of step 212 and reselect the update package.

  In addition, the hardware allocation value (CPU allocation rate) allocated to each virtual machine (each OS) in the performance information of the virtual machine so that the update end timings coincide or become a time difference within a predetermined range. The predicted time of each process may be adjusted by adjusting the disk I / O bandwidth and the network bandwidth (R1, R3, R4)). In this case, the predicted time of the first OS is also an adjustment target.

  In addition, after calculating the estimated time of each process for the selected update package, the schedule determination for each process is suspended, and the next update package is selected to calculate the estimated time for each process of the update package After performing the process first, the schedule of each process of the plurality of update packages may be determined collectively. For example, instead of the order of “download of update package 1 → update of update package 1 → download of update package 2 → update of update package 2”, as shown in FIG. 10, “download of update package 1 → update of update package 2” “Download → update of update package 1 → update of update package 2”. In this way, whether to change the order across update packages may be added to the attribute information of the update package. Then, if the hypervisor refers to the attribute information and the order change is permitted, the schedule may be determined by changing the order as described above.

  Note that hardware resource allocation may be performed as follows. As shown in FIG. 10, during the period T1 in which updates are mainly performed on the first OS side and downloads are mainly performed on the second OS side, the CPU and HDD loads on the first OS side are large, and the network load is large on the second OS side. The allocation ratio of the CPU and HDD to the first OS is higher than that of the second OS, and the allocation ratio of the network interface to the second OS is higher than that of the first OS. In the next period T2, download is mainly executed on the first OS side, and updates and database initialization are mainly performed on the second OS side. Therefore, the network load is large on the first OS side, and the CPU and HDD are on the second OS side. Since the load is large, the allocation ratio of the network interface to the first OS is set higher than that of the second OS, and the allocation ratio of the CPU and HDD to the second OS is set higher than that of the first OS.

  After the scheduling process is completed as described above, the process proceeds to step 106 in FIG.

  In step 106, the OS of each virtual machine is controlled so that the update process by each update package is executed according to the determined schedule, and hardware resources are allocated according to the determination of the allocation rate.

  If it is expected that the time required for each process will not be the predicted time calculated above during the actual execution of the update process, each of the process times will match the predicted time during the execution of the update process. In addition, the determined CPU allocation rate, disk I / O bandwidth, and network bandwidth may be dynamically adjusted.

  Furthermore, the rescheduling process may be performed after the update process is started. In the rescheduling process, the allocation state of the hardware resources of each virtual machine at the start of execution of the rescheduling process is acquired, and the scheduling process of step 104 is executed for the unexecuted update package and each process. When a new schedule and allocation rate are determined in this way, thereafter, each OS is controlled to execute an unexecuted process according to the newly determined schedule, and hardware resource allocation is also new. Do it according to the decision. Note that rescheduling may be performed at predetermined time intervals in order to avoid a load on hardware resources due to frequent rescheduling processing.

  Furthermore, in the present embodiment, the image forming apparatus 10 that implements a virtual system by allocating hardware resources using a hypervisor has been described as an example. However, the present invention is not limited to this. For example, FIG. As shown in FIG. 4, instead of the hypervisor, a host OS that manages the allocation of hardware resources is provided separately from the plurality of OSs, and the host OS performs scheduling processing for updating the plurality of OSs. Similarly to the above, the update of each OS may be controlled.

  In the present embodiment, the image forming apparatus is described as an example in which the update control apparatus is applied. However, the present invention is not limited to this, and the present invention is not limited to this and may be applied to various home appliances and office equipment. Applicable.

10 image forming apparatus 12 apparatus control unit 12A CPU
12B Memory 12C Storage unit 14 Image reading unit 16 Image forming unit 18 Operation panel 18B Operation receiving unit 18A Display unit 20 Network interface 24 Bus

Claims (3)

An update control device provided in an information processing apparatus in which a plurality of operating systems operate by sharing hardware resources,
Based on each of update software attribute information for updating each of the plurality of operating system software and each of hardware resource information allocated to each of the plurality of operating systems, A predicting means for predicting the time required for each process executed when updating by the update software;
Compared with the case where the plurality of operating systems start executing the update software all at once and update the software of each operating system based on the predicted time, the processing between the plurality of operating systems is performed. First determination means for determining the execution order and execution start timing of each of the processes executed during the update by the update software so that the time during which the processes having the same contents are executed in parallel is shortened;
Second determining means for determining an allocation rate of the hardware resources for each of the plurality of operating systems according to the predicted time, the execution order, and the execution start timing;
Control means for performing control so that each process by the update software is executed in the determined execution order and execution start timing, and the hardware resources are allocated at the determined allocation rate;
An update control device. The control means, during execution of each process by the update software, prediction by the prediction means, determination of the execution order and execution start timing by the first determination means, for each predetermined time interval, The allocation rate is determined by the second determination means, and each process by the update software is executed in the execution order and execution start timing newly determined by the control, and newly determined by the control. The update control device according to claim 1, wherein control is performed so that the hardware resource is allocated at a predetermined allocation rate. A computer provided in an information processing apparatus in which a plurality of operating systems operate by sharing hardware resources;
Based on each of update software attribute information for updating each of the plurality of operating system software and each of hardware resource information allocated to each of the plurality of operating systems, A predicting means for predicting the time required for each process executed when updating by the update software;
Compared with the case where the plurality of operating systems start executing the update software all at once and update the software of each operating system based on the predicted time, the processing between the plurality of operating systems is performed. First determination means for determining the execution order and execution start timing of each of the processes executed during the update by the update software so that the time during which the processes having the same contents are executed in parallel is shortened;
Second determining means for determining an allocation rate of the hardware resources for each of the plurality of operating systems according to the predicted time, the execution order, and the execution start timing; and the determined execution order and execution Control means for performing control so that each processing is executed at a start timing and the hardware resources are allocated at the determined allocation rate;
Program to function as.

Images