摘要:
A multiprocessor system includes a number of sub-processor systems, each substantially identically constructed, and each comprising a central processing unit (CPU), and at least one I/O device, interconnected by routing apparatus that also interconnects the sub-processor systems. A CPU of any one of the sub-processor systems may communicate, through the routing elements, with any I/O device of the system, or with any CPU of the system. Communications between I/O devices and CPUs is by packetized messages. Interrupts from I/O devices are communicated from the I/O devices to the CPUs (or from one CPU to another CPU) as message packets. CPUs and I/O devices may write to, or read from, memory of a CPU of the system. Memory protection is provided by an access validation method maintained by each CPU in which CPUs and/or I/O devices are provided with a validation to read/write memory of that CPU, without which memory access is denied.
摘要:
Method and system of determining whether a user program has made a system level call and thus whether the user program is uncooperative with fault tolerant operation. Some exemplary embodiments may be a processor-based method comprising providing information from a first processor to a second processor (the information indicating that a user program executed on the first processor has not made a system level call in a predetermined amount of time), and determining by the first processor, using information from the second processor, whether a duplicate copy of the user program substantially simultaneously executed in the second processor has made a system level call in the predetermined amount of time.
摘要:
A method and system of implementing a persistent memory. At least some of the illustrative embodiments are a system comprising a first computer slice comprising a memory, a second computer slice comprising a memory (the second computer slice coupled to the first computer slice by way of a communication network at least partially external to each computer slice), and a persistent memory comprising at least a portion of the memory of each computer slice (the portion of the memory of the first computer slice storing a duplicate copy of data stored in the portion of the memory of the second computer slice). The persistent memory is accessible to an application program through the communication network.
摘要:
Adaptive sets of lanes are configured between routers in a system area network. Source nodes determine whether packets may be adaptively routed between the lanes by encoding adaptive control bits in the packet header. The adaptive control bits also facilitate the flushing of all lanes of the adaptive set. Adaptive sets may also be used in uplinks between levels of a fat tree.
摘要:
A multiprocessor system includes a number of sub-processor systems, each substantially identically constructed, and each comprising a central processing unit (CPU), and at least one I/O device, interconnected by routing apparatus that also interconnects the sub-processor systems. A CPU of any one of the sub-processor systems may communicate, through the routing elements, with any I/O device of the system, or with any CPU of the system. The CPUs are structured to operate in one of two modes: a simplex mode in which the two CPUs operate independently of each other, and a duplex mode in which the CPUs operate in lock-step synchronism to execute each instruction of identical instruction streams at substantially the same time. Communications between I/O devices and CPUs is by packetized messages. Interrupts from I/O devices are communicated from the I/O devices to the CPUs (or from one CPU to another CPU) as message packets. CPUs and I/O devices may write to, or read from, memory of a CPU of the system. Memory protection is provided by an access validation method maintained by each CPU in which CPUs and/or I/O devices are provided with a validation to read/write memory of that CPU, without which memory access is denied.
摘要:
A dual processor computer system includes a first processing system having a central processing unit which executes a series of data processing instructions, a data bus system for transferring data to and from the first central processing unit, a memory unit coupled to the first central processing unit, and a cross-link communications element for transferring data into and out of the first processing system. A similarly configured second processing system, operating independently of the first processing system, is also provided. The cross-link communications element associated with the second processing system is coupled to the cross-link communication element of the first processing system, for transferring data into the second processing system from the first processing system and for transferring data into the first processing system from the second computer system.
摘要:
A data transfer system for use in transferring data between a memory and an input/output system in a digital data processing system. The data transfer system includes a plurality of buffers into which data can be loaded from the memory or the input/output system. A buffer control selects the buffer to be loaded, and control signals from the memory govern the transfer of data from the memory into and out of the selected buffer.
摘要:
A central processor unit for a digital data processing system that processes prefetched instructions including a conditional branch instruction. The processor includes a fetch unit that has separate portions, one that retrieves operands and the other that retrieves instructions. When the fetch unit fetches a conditional branch instruction, it may continue to prefetch "branch not taken" instructions using the instruction fetch portion. The fetch unit initially uses the operand fetch portion to prefetch "branch taken" instructions. If it is determined that the branch is not taken, the prefetch operation is aborted, otherwise the prefetch operation is allowed to continue to provide the next instruction used by the processor.
摘要:
A dual processor data processing system having interprocessor error checking includes a first central processing unit executing a series of instructions. A second central processing unit executes the same series of instructions independently of and in synchronism with the first central processing unit. A first data bus is coupled to the first central processing unit for receiving data to be input to the first central processing unit and a second data bus is coupled to the second central processing unit for receiving data to be input to the second central processing unit. Error checking devices are coupled to the first and second data busses for checking data transmitted over the first and second data busses and for detecting errors on I/O reads prior to delivery of the data to the first and second central processing units. The error checking devices include comparison means for indicating an error when the data on the first and second data busses are unequal. Error isolation devices are responsive to an error detected from the error checking means for analyzing the cause of error while maintaining system synchronization.
摘要:
A data handling system for transferring data between two units, the data being transferred in blocks of a selected number of data words, up to predetermined maximum number. A buffer stores the data being transferred. The buffer includes a plurality of stages arranged serially from an input end to an output end, the number of stages being equal in number to the predetermined maximum number of data words that may be transferred in a block. If the number of data words being transferred is less than the predetermined maximum number, as indicated by a control signal from the unit transmitting the data, the buffer either receives the data in the stage a number of stages from the output end, or transmits the data from the stage a number of stages from the input end, equal to the number of words being transferred in the block.