摘要:
In a computer system including a job scheduler which divides an input job and determines a client apparatus to assign the divided job, and a task manager which transmits the divided job to the client apparatus determined by the job scheduler, the job scheduler assigns the same job to a plurality of client apparatuses when the number of client apparatuses is larger than the number of divided jobs, and when the job is finished in one of the plurality of client apparatuses assigned the same job, execution of the job in the remaining client apparatuses is canceled so that the time until all jobs are finished is shortened.
摘要:
In a computer system including a job scheduler which divides an input job and determines a client apparatus to assign the divided job, and a task manager which transmits the divided job to the client apparatus determined by the job scheduler, the job scheduler assigns the same job to a plurality of client apparatuses when the number of client apparatuses is larger than the number of divided jobs, and when the job is finished in one of the plurality of client apparatuses assigned the same job, execution of the job in the remaining client apparatuses is canceled so that the time until all jobs are finished is shortened.
摘要:
Even when a CPU gains higher performance and a distributed computing environment is being put in place, it is not easy to follow the speed of hardware image processing implemented by dedicated hardware. Hence, jobs of filter processing having a heavy operation load and low distributability are selectively and intensively input to a small number of resources having performance as high as possible. On the other hand, since the color conversion processing and gamma processing have high distributability, even when jobs are input to many resources having low performance, an improvement of the processing speed can be expected. Therefor, color conversion processing and gamma processing distribute jobs to many resources.
摘要:
When the distributed load system of grid computing is applied to printers, multi-functional peripheral equipments, and the like connected to a computer network, if a device which inputs a job received from the client is executing processing, the distributed processing requires much time, and the user of the client cannot obtain the print result soon. Hence, when a print job is input, the processing time upon executing distributed processing of data processes in the print job by computer devices connected to the computer network is estimated. Based on the estimation result, the print job or jobs divided from the print job are transmitted to the computer devices connected to the computer network.
摘要:
A mechanism that attains high-speed processing by distributing a great deal of processing to a plurality of devices upon processing does not perform processing distribution which considers hardware resources of devices used in distributed processing. When rendering processing of a print job is executed by distributed processing using devices connected to a computer network, it is determined whether hardware or software rendering processing is executed (S1102, S1103). Image processing devices which can implement hardware or software rendering processing are sought from devices connected to the computer network (S1104, S1106). Based on the determination and seek results, devices to which jobs divided from the print job are transmitted are notified (S1108).
摘要:
In a network print system in which a printer and at least one client are connected to each other through a network, and at least a part of a job for the printer can be subjected to load distributed processing by grid computing, wherein when a printer driver of the printer is installed on any one of the at least one client, the system is adapted to build a grid computing network. In this arrangement, it is easily to build the grid computing network in which the printer job is subjected to load distributed processing.
摘要:
In a network print system in which a printer and at least one client are connected to each other through a network, and at least a part of a job for the printer can be subjected to load distributed processing by grid computing, wherein when a printer driver of the printer is installed on any one of the at least one client, the system is adapted to build a grid computing network. In this arrangement, it is easily to build the grid computing network in which the printer job is subjected to load distributed processing.
摘要:
Even when a CPU gains higher performance and a distributed computing environment is being put in place, it is not easy to follow the speed of hardware image processing implemented by dedicated hardware. Hence, jobs of filter processing having a heavy operation load and low distributability are selectively and intensively input to a small number of resources having performance as high as possible. On the other hand, since the color conversion processing and gamma processing have high distributability, even when jobs are input to many resources having low performance, an improvement of the processing speed can be expected. Therefor, color conversion processing and gamma processing distribute jobs to many resources.
摘要:
When the distributed load system of grid computing is applied to printers, multi-functional peripheral equipments, and the like connected to a computer network, if a device which inputs a job received from the client is executing processing, the distributed processing requires much time, and the user of the client cannot obtain the print result soon. Hence, when a print job is input, the processing time upon executing distributed processing of data processes in the print job by computer devices connected to the computer network is estimated. Based on the estimation result, the print job or jobs divided from the print job are transmitted to the computer devices connected to the computer network.
摘要:
When a target device of distributed processing has shifted to a power save mode, if the target device is resumed to the normal operation to attain the distributed processing, resources of the target device which are not required for the distributed processing waste electric power. Therefore, when a print job to be processed by grid processing is received, resources required for the grid processing from resources in a waiting state, and the grid processing is executed to process the received print job.