摘要:
In an information processing system having plural processing units connected through a communication medium, each processing unit operates to receive information to be transmitted from a processing portion operating in the processing unit to the communication medium and a content code predetermined according to the content of the information to be transmitted. Based on the received content code, the processing unit operates to determine an address of the destination processing unit, a sending mode used for transmitting the information, and whether or not the more reliable communication method arranged to establish a connection between the processing units is used for transmitting the information. Then, by using the sending mode and the communication method obtained as the determined result, the information is transmitted to the processing unit having an address obtained as the result determined through the communication medium.
摘要:
In a distributed processing system including a plurality of programs distributed in a plurality of processing units and in which a program on reception side performs processing using a message transmitted from a program on transmission side, a management table for managing start of programs on reception side is provided in a processing unit in which the program on reception side is provided. A program utilizing the message transmitted from the transmission side is recognized and the recognized program is started at the start timing indicated in the management table. The start timing of the program set in the management table may be registered previously or the message may include information indicating the start timing.
摘要:
In a distributed processing system including a plurality of programs distributed in a plurality of processing units and in which a program on reception side performs processing using a message transmitted from a program on transmission side, a management table for managing start of programs on reception side is provided in a processing unit in which the program on reception side is provided. A program utilizing the message transmitted from the transmission side is recognized and the recognized program is started at the start timing indicated in the management table. The start timing of the program set in the management table may be registered previously or the message may include information indicating the start timing.
摘要:
A method for executing programs in a distributed processing system includes executing in a plurality of first processors different programs which perform the same function. Messages, containing data representing execution result output from said different programs and attribute information indicating function to be performed by the different programs, are transmitted from the first processors to a network. A second processor then receives these messages and selects a message therefrom based on an analysis of the attribute information contained in all of the received messages. The second processor then executes an application program using the data contained in the selected message. Alternatively, the message-selecting step is made based on both the attribute information and data contained in the messages received in the second processor. Each of the messages from the first processors may include graphic image data which can be analyzed and shown on a display.
摘要:
A data management method in a distributed processing system having a plurality of processors connected through a common transmission line with at least one processor having at least one external device connected therewith, for sending to the common transmission line access commands issued to the external device by programs asynchronously executed in the programs as access messages; and wherein the processor having the external device connected therewith takes in any of the access messages related to the external device connected therewith from among the access messages asynchronously sent to the common transmission line by the programs, and the processor matches order of access to the external devices connected therewith among the received access messages.
摘要:
A resource management method and system for version-up in a multiprocessor system having functions that a processor to be made version-up investigates resources under control by the own processor itself before deciding if any of the resources are to be moved and/or duplicated, that the processor can decide the processors to which the movement and/or duplication of any of the resources is destined, that the resource is moved or duplicated to the destined processor. After that, version-up of the processor to be made version up is executed, thereby effectively using the resources and making version-up while the multiprocessor system is under operation.
摘要:
In a multi-computer system in which a plurality of computers connected to each other through a transmission medium share a process to be carried out under relationships among a method performed by these computers executing a control instruction, characterized as; while the plurality of computers each have specific control instruction storage means for storing instructions having influence on normal operations of the multi-computer system, when any of the plurality of computers defects an input instruction corresponding to one stored in the specific control instruction storage means, determines the status of the shared process being carried out by other computers in connections to the self-computer, and uses the result of the determination in the execution of the detected control instruction.
摘要:
A replicated data processing method and apparatus for a distributed processing system are provided in which a plurality of processors are connected to one another through a common transmission line. The processors perform processes using replicated data produced by resources such as files replicated in respective processors. Any processor issues an access request and broadcasts the request to the common transmission line. User's program of the processor is initiated by a first-received data in response to the request, and checks matching between the first-received data of the replicated data and the subsequently received data to repeat a retry processing for the request until a mismatch is removed.
摘要:
According to the present invention, a file recovery system is provided in which the commands for recovering files such as a historical file are unnecessitated and the recovery system configuration is independent of the host processing system. That is, since the equipment for achieving a file recovery does not have information about the location of the recovery data, information about the content of a file to be recovered is transmitted to a transmission line. The other equipments asynchronously receive the information and judges whether or not a file containing the same content exists in the own system. If its file exists, the content of the file is transmitted to the transmission line. The equipment achieving the file recovery asynchronously receives from the transmission line the file data of files such as a historical file and establishes data consistency by use of the buffered current data and the event number, thereby recovering files such as a historical file.
摘要:
Several processors and several communication control units are connected to a transmission medium. Each communication control unit has one or more terminal devices connected to it, and so any terminal device can be coupled to any processor. Each processor includes a central processing unit, a memory, and a control manager. The memory stores first information regarding processing being done with a terminal device coupled to the processor and second information regarding the communication control units connected to each of the processors and regarding the terminal devices being serviced by the processor. The control manager manages the communication control units based on the stored information. Information about the connections between the processors and the communication control units is broadcast to all processors. If any processor is faulty, another processor can take over the processing being done by the faulty processor. Likewise, further, processors can be added to the network, and their connection status is known to all other processors.