US 2006/0150006 A1Published application

Securing time for identifying cause of asynchronism in fault-tolerant computer

AssigneeNEC Corporation

InventorFumitoshi Mizutani

Life of this patent

Estimating the term…

Loading patent…

Abstract

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.

Claims8

Dependent claims

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 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 its own 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 as to a writing area for storing data in the main storage unit of its own system each time data is stored in the main storage unit of its own 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, whose writing area is stored in said trace memory, 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 predetermined time is reached.

5 more elements

A method of controlling a fault-tolerant computer having a pair of duplex systems having respective CPU subsystems each including a CPU and a main storage unit, said CPU subsystems being operable identically in lock-step synchronism, comprising the steps of:

a) in each of said duplex systems, continuously operating the CPU of its own 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 as to a writing area for storing data in the main storage unit of its own system each time data is stored in the main storage unit of its own system after the asynchronous operation is detected;
and
c) in each of said duplex systems, if the asynchronous operation is detected, holding a DMA transfer process to transfer the data, whose writing area is stored in said trace memory, 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 predetermined mined time is reached.

1 more element

Dependencies are inferred from the claim text.