초록 ▼

A fault-tolerant computer has a pair of duplex systems having respective CPU subsystems that are operable identically in lock-step synchronism. Each of the duplex systems has a CPU, a main storage unit, a CPU bus controller, and a DMA controller. The CPU and the main storage unit are included in eac

A fault-tolerant computer has a pair of duplex systems having respective CPU subsystems that are operable identically in lock-step synchronism. Each of the duplex systems has a CPU, a main storage unit, a CPU bus controller, and a DMA controller. The CPU and the main storage unit are included in each of the CPU subsystems. The CPU bus controller continuously operates the CPU of its own system even if it detects an asynchronous operation while the CPU subsystems are operating in synchronism with each other. Even if the asynchronous operation is detected, the DMA controller holds a DMA transfer process for transferring data stored in the main storage unit of its own system or the other system to the main storage unit of the other system or its own system after the asynchronous operation is detected until a certain time is reached.

대표청구항 ▼

What is claimed is: 1. A fault-tolerant computer comprising: a pair of duplex systems having respective CPU subsystems operable identically in lock-step synchronism; each of said duplex system comprising: a CPU of the duplex system included in one of said CPU subsystems; a main storage unit include

What is claimed is: 1. A fault-tolerant computer comprising: a pair of duplex systems having respective CPU subsystems operable identically in lock-step synchronism; each of said duplex system comprising: a CPU of the duplex system included in one of said CPU subsystems; a main storage unit included in said one of the CPU subsystems; a CPU bus controller for continuously operating the CPU of the duplex system without shutdown if an asynchronous operation is detected while said CPU subsystems are operating in synchronism with each other; a trace memory for storing information of a writing area for storing data in the main storage unit of the duplex system each time data is stored in the main storage unit of the duplex system after the asynchronous operation is detected by said CPU bus controller; a DMA controller for, if the asynchronous operation is detected by said CPU bus controller, holding a DMA transfer process to transfer the data having the writing area stored in said trace memory, the data being stored in the main storage unit of the duplex system or the other duplex system, the DMA transfer process to transfer the data to the main storage unit of the other duplex system or the duplex system, after the asynchronous operation is detected until a predetermined time is reached; a buffer for temporarily storing input data and thereafter outputting the input data; and an access comparator for sending an interrupt signal to said CPU of the duplex system if details of an access from the CPU of the duplex system through said buffer and details of an access from the CPU of the other duplex system are compared with each other and detected as being not in conformity with each other and also if the storage capacity of said buffer exceeds a predetermined value; wherein said predetermined time to be reached in said DMA controller is a time when the interrupt signal is sent from said access comparator to said CPU of the duplex system. 2. A fault-tolerant computer according to claim 1, wherein said CPU bus controller detects the asynchronous operation by monitoring an operating state of a CPU bus of the duplex system, sending CPU bus operational information representative of the monitored operating state to the CPU bus controller of the other duplex system, and comparing the CPU bus operational information of the duplex system and the CPU bus operational information of the other duplex system with each other. 3. A fault-tolerant computer comprising: a pair of duplex systems having respective CPU subsystems operable identically in lock-step synchronism; each of said duplex systems comprising: a CPU of the duplex system included in one of said CPU subsystems; a main storage unit included in said one of the CPU subsystems; a CPU bus controller for continuously operating the CPU of the duplex system without shutdown if an asynchronous operation is detected while said CPU subsystems are operating in synchronism with each other; a trace memory for storing information of a writing area for storm data in the main storage unit of the duplex system each time data is stored in the main storage unit of the duplex system after the asynchronous operation is detected by said CPU bus controller; and a DMA controller for, if the asynchronous operation is detected by said CPU bus controller, holding a DMA transfer process to transfer the data having the writing area stored in said trace memory, the data hem stored in the main storage unit of the duplex system or the other duplex system, the DMA transfer process to transfer the data to the main storage unit of the other duplex system or the duplex system, after the asynchronous operation is detected until a predetermined time is reached, wherein said DMA controller sends an interrupt signal to said CPU of the duplex system if the storage capacity of said trace memory exceeds a predetermined value, and said predetermined time to be reached in said DMA controller is a time when the interrupt signal is sent from said DMA controller to said CPU of the duplex system. 4. A fault-tolerant computer according to claim 3, wherein said CPU bus controller detects the asynchronous operation by monitoring an operating state of a CPU bus of the duplex system, sending CPU bus operational information representative of the monitored operating state to the CPU bus controller of the other duplex system, and comparing the CPU bus operational information of the duplex system and the CPU bus operational information of the other duplex system with each other. 5. A method of controlling a fault-tolerant computer having a pair of duplex systems having respective CPU subsystems each duplex system including a CPU of the duplex system and a main storage unit, said CPU subsystems being operable identically in lock-step synchronism, said method comprising the steps of: a) in each of said duplex systems, continuously operating the CPU of the duplex system without shutdown if an asynchronous operation is detected while said CPU subsystems are operating in synchronism with each other; b) in each of said duplex system, storing, in a trace memory, information of a writing area for data in the main storage unit of the duplex system each time data is stored in the main storage unit of the duplex system after the asynchronous operation is detected; c) in each of said duplex systems, if the asynchronous operation is detected, holding a DMA transfer process to transfer the data having the writing area stored in said trace memory, the data being stored in the main storage unit of the duplex system or the other duplex system, the DMA transfer process to transfer the data to the main storage unit of the other duplex system or the duplex system, after the asynchronous operation is detected until a predetermined time is reached; and d) in each of said duplex systems, sending an interrupt signal to said CPU of the duplex system if details of an access from the CPU of the duplex system through a buffer and details of an access from the CPU of the other duplex system are compared with each other and detected as being not in conformity with each other and also if the storage capacity of said buffer exceeds a predetermined value; wherein, in said step c), said predetermined time is a time when the interrupt signal is sent to said CPU of the duplex system. 6. A method according to claim 5, wherein said step a) comprises the steps of, in each of said duplex systems, monitoring an operating state of a CPU bus of the duplex system, sending CPU bus operational information representative of the monitored operating state to the other duplex system, and comparing the CPU bus operational information of the duplex system and the CPU bus operational information of the other duplex system with each other. 7. A method of controlling a fault-tolerant computer having a pair of duplex systems having respective CPU subsystems each duplex system including a CPU of the duplex system and a main storage unit, said CPU subsystems being operable identically in lock-step synchronism, said method comprising the steps of: a) in each of said duplex systems, continuously operating the CPU of the duplex system without shutdown if an asynchronous operation is detected while said CPU subsystems are operating in synchronism with each other; b) in each of said duplex systems, storing, in a trace memory, information of a writing area for storm data in the main storage unit of the duplex system each time data is stored in the main storage unit of the duplex system after the asynchronous operation is detected; c) in each of said duplex systems, if the asynchronous operation is detected, holding a DMA transfer process to transfer the data having the writing area stored in said trace memory, the data being stored in the main storage unit of the duplex system or the other duplex system, the DMA transfer process to transfer the data to the main storage unit of the other duplex system or the duplex system, after the asynchronous operation is detected until a predetermined time is reached; and d) in each of said duplex systems, sending an interrupt signal to said CPU of the duplex system if the storage capacity of said trace memory exceeds a predetermined value; wherein, in said step c), said predetermined time is a time when the interrupt signal is sent to said CPU of the duplex system. 8. A method according to claim 7, wherein said step a) comprises the steps of, in each of said duplex systems, monitoring an operating state of a CPU bus of the duplex system, sending CPU bus operational information representative of the monitored operating state to the other duplex system, and comparing the CPU bus operational information of the duplex system and the CPU bus operational information of the other duplex system with each other.