Abstract:
A method and apparatus for controlling data flow between a computer (10) and a group of memory devices (18A...18M) arranged in a particular logical configuration. The system includes a group of first level controllers (12A, 12B) and a group of second level controllers (14A, 14B). The first level controllers (12A, 12B) and the second level controllers (14A, 14B) work together such that if one of the second level controllers (14a, 14B) fails, the routing between the first level controllers (12A, 12B) and the memory devices (18A...18M) is switched to a properly functioning second level controller (14A, 14B) without the need to involve the computer (10) in the rerouting process. The logical configuration of the memory devices (18A...18M) remains constant. The invention also includes switching circuitry (16A1...16M6) which permits a functioning second level controller (14A, 14B) to assume control of a group of memory devices (18A...18M) formerly controlled by the failed second level controller (14A, 14B).
Abstract:
An improved interface system based in part on the SCSI standard is provided. A single cable data bus simultaneously transfers several bytes of information between two devices. The interface system transfers multiple-byte commands, messages, status information or data in a single parallel transfer. A microsequencer is provided to permit data transfers across the interface without requiring burdensome attention from a processor in a device involved in the transfer.
Abstract:
A circuit for decoding a high speed Manchester encoded digital communication signal is provided. The circuit includes a pair of latch circuits which are used to detect clock edges in the encoded signal for providing respectively set and reset pulses to a third latch circuit, an output of which comprises the decoded data of the Manchester code signal. Additional logic is provided to extract a clock signal from the Manchester code signal.
Abstract:
A scheduling mechanism is provided for controlling when the arbitration circuit of a node (10) sharing a CSMA communication medium (14) is to start CSMA arbitration for access to the communication medium once the node has a message ready for transmission, the scheduling mechanism delaying the arbitration circuit (100) from seeking access if total transmission activity (TCU) on the communication medium exceeds a total use threshold (TMU) value and transmission activity (LCS) of the node exceeds a local use threshold value (LMS), and otherwise permitting the arbitration circuit to seek access to the communication medium by arbitration in accordance with a priority value assigned to the node.
Abstract:
A network-type (10) data processing system is provided. The system can support multiple simultaneous exchanges of data, and includes multi-port storage devices (100-104) in which all ports can be active at all times. On initialization of the system, each storage device can announce itself through all of its ports simultaneously through at least 2 paths.
Abstract:
A method and apparatus for transferring data from one device interface to another device interface via elements of a staging memory and a direct memory access (DMA) channel.
Abstract:
The present invention provides a method and apparatus for dynamically modifying the priority of access to a bus (16), where the bus (16) has control and arbitration functions distributed among the devices (20) coupled to the bus (16), with each device (20) having a fixed priority level. Access to the bus by particular devices is selectively inhibited, preventing them from asserting their fixed priority level. In a preferred embodiment, the present invention provides control over the reselection of a SCSI bus (16) by a plurality of SCSI devices (20) coupled to the bus (16) by providing a pseudo busy signal to SCSI devices (20) from which reselection is not desired. In this fashion, an initiator may issue a plurality of commands to the SCSI devices (20) and control the order in which the devices (20) will be serviced when ready.
Abstract:
A computing system providing resource management in a multiple resource system. In the preferred embodiment, the computing system has a plurality of resources (106, 108, 138a, 140a, 142a, 126a-126m) for storing, transmitting or manipulating data. The system also has a fault management subsystem (122a, 150, 152) that accesses and operates the resources when the resources are in a first availability state and an operational subsystem for accessing and operating the resources when the resources are in a second availability state. The system has a mechanism (200, 118, 152) for providing real time sharing of any of the resources between the fault management subsystem (122a, 150, 152) and the operational subsystem without disrupting the services provided by the operational subsystem. In addition, the system has a mechanism for representing the operational interdependencies of the resources by organizing the resources in a logical structure in which each resource is a node conceptually connected to at least one other resource.
Abstract:
A mass memory system for digital computers is disclosed. The system has a plurality of disk drives (250-255) coupled to a plurality of small buffers (240-245). An Error Correction Controller (260 and 270) is coupled to a plurality of X-bar switches (210-215), the X-bar switches being connected between each disk drive and its buffers. Data is read from and written to the disk drives in parallel and error correction is also performed in parallel. The X-bar switches are used to couple and decouple functional and nonfunctional disk drives to the system as necessary. Likewise, the buffers can be disconnected from the system should they fail. The parallel architecture, combined with a Reed-Solomon error detection and correction scheme and X-bar switches allows the system to tolerate and correct any two failed drives, allowing for high fault-tolerance operation.
Abstract:
A method and apparatus for identifying members of a set of physical mass storage devices acting as one logical mass storage device are provided. Each physical mass storage device is assigned a membership signature identifying it as a valid member of the set. Whenever a member of a set undergoes a membership state-changing event, the membership signatures of all other devices in the set are changed, so that the member with the changed membership state no longer has a valid signature. When the member is reinstalled, it can be given a new valid signature after it is uptdated or regenerated.