Digital Equipment Corporation

Hewlett-packard Development Company, L.p.

Other

Hewlett Packard Enterprise Development LP

Edward A Gardner

Barry L Rubinson

Richard Fitzhugh Wrenn

Richard Perham Helliwell

Robert G Bean

Richard Franklin Lary

Peter T McLean

Michael E Beckman

O Winston Sergeant

David William Thiel

Robert C Frame

Fernando A Zayas

Robert E Willard

Dale R Keck

Verell Boaen

William Watson

Catharine Van Ingen

Robert Blackledge

Mark A Parenti

Michael Chow

William A Grace

A transmitting end-point computes a current transmission rate for each respective outbound half-route of outbound half-routes of a route set between transmitting and receiving end-points. The transmitting end-point receives, from the receiving end-point via a respective inbound half-route of the route set, a transmission rate limit for each respective outbound half-route, the transmission rate limit computed by the receiving end-point from routing headers of messages received by the receiving end-point on the respective outbound half-route, wherein the transmission rate limit for each respective outbound half-route places an upper bound on the current transmission rate for transmissions issued on the respective outbound half-route.

System and method for message routing in a network

A system and method for message routing in a computer network. The present system establishes a routing layer between the network and transport layers in a network, for communicating routing information between end-points in the network. A route management connection for communicating route management information between two ports in the network is established between pseudo end-points in the network. A route and route set are established between end-points in the network, where the route set provides a relationship between two end-point incarnations that indicates the existence of at least one route between the end-points. A route set management connection is established between the end-points for managing connections that are established on the route set. A route between the end points is selected using a route in the route set, and an application connection is established between two applications running on respective end-points in the network, using the route set and the route set management connection.

A system and method for selecting a route, from within the set of network routes between the source and destination end-points, for retransmission of information initially sent via a suspected failed route such that the selected route has a minimum number of network components in common with the suspected failed route. In one aspect, the routes between source and destination end-points are grouped into a route set and placed in order in a circular list by the destination end-point. When the source end-point deems it necessary to retransmit information sent via a suspected failed route, it selects the next acceptable route following the suspected failed route from the circular list. The circular list order is determined such that adjacent routes have few common network components.

System for selecting routes for retransmission in a network

A method for controlling command message flow in a network including a server and a client. A command window, comprising a maximum number of command messages that may be outstanding at the server, is included in messages sent from the server to the client. The value of the command window at the server is modified in accordance with available server resources for receiving command messages. When there are insufficient resources at the server to process one of the command messages delivered to the server, then a pause message is sent to the client indicating which said command message cannot be received; indicia is stored that indicates the command message initially discarded; and subsequent said command messages delivered to the server are discarded until an initially discarded said command message is again delivered to the server.

Controlling command message flow in a network

A system for addressing end-points in a network wherein a transmission is sent between a source end-point and a destination end-point via a respective source port and a destination port connected by a fabric. A payload offset and buffer pool handle are included in a network header sent with the transmission. The buffer pool handle is used to determine a specific queue of pool buffers within the destination port. A pool buffer descriptor, describing a header buffer and a payload buffer each within the destination end-point, is dequeued from a queue indicated by the buffer pool handle. The payload offset in the network header is then decoded to locate a first number of bytes of data in the transmission to be transferred into the header buffer a second number of bytes of data to be transferred into the payload buffer.

System for communicating between network end-points using a payload offset and buffer pool handle

A system and method for determining route quality in a network in which multiple routes exist between a transmitting end-point and a receiving end-point. Initially, a routing sequence number is associated with each transmission sent on a route in the network, such that consecutive transmissions sent on one of the routes have consecutive routing sequence numbers. The routing sequence number for a first one of the routes is included in transmissions sent on a second one of the routes. The routing sequence number of the first route, received on the second route, is used as an indicator of the route quality of the first route, and this indicator is compared to an expected routing sequence number for the first route to determine a likelihood of whether the transmissions sent via the first route are either lost or delayed. The expected routing sequence number is normally incremented each time a transmission is received on the second route.

System for determining network route quality using sequence numbers

A method and apparatus for transferring data among a plurality of sending and receiving nodes is disclosed. All nodes use the same control lines in the same manner. In a first phase all nodes assert a first control line and transfer data to the data bus concurrent with the assertion of a second control line. After each of the nodes has released the first control line, a second phase occurs with all nodes reading the data. The data transfer is complete when a preconditioned second control line is released by all of the nodes. A third phase resets the control lines to enable the next data transfer cycle to begin. Nodes may take a variable time for each phase of a data transfer and the transfer rate is automatically adjusted to the slowest node.

Asynchronous data transfers among a plurality of computer devices

A communication interface between a port driver and an port adapter of a host computer includes a singly-linked queue resident in a host memory of the computer. The queue includes a header element and a stopper element with message entries linked therebetween. The message entries contain information to be exchanged between the port driver and port adapter. Each message entry includes at least a carrier that contains a forward link pointer to the next entry in the queue and message/response specific control information. The stopper element identifies the end of the queue and is distinguishable from a message entry by a valid indicator, e.g., the state of at least one bit of the forward link pointer. Functionally, the stopper entry allows the port driver and port adapter to concurrently insert and remove message entries of the singly-linked queue in a reliable manner without the use of any external synchronization mechanism.

Queue apparatus and mechanics for a communications interface architecture

A method and apparatus for selecting a particular node from a plurality of nodes connected to a common bus to allow the node to use the bus. The nodes have a pre-determined priority. After initially enabling the nodes, the bus is monitored for a bus idle condition. It is then determined which of the nodes are enabled message nodes, which are enabled nodes that have a message to send on the bus. There is then arbitration between the enabled message nodes after the bus is in the bus idle condition for a first period of time, such that the enabled message node having the highest pre-determined priority among the enabled message nodes is disabled for arbitration purposes, and also at the same time selects a target and performs a transfer. This procedure is repeated until all the enabled message nodes have been disabled. Thereafter, all of the nodes on the bus are enabled when the bus is in the bus idle condition for a second period of time, which is longer than the first period of time.

Fair arbitration scheme for arbitrating between multiple nodes in a

A secondary storage facility having a drive and a controller employing multiple error recovery techniques; the controller signals the drive to try such techniques in sequence, according to descending a priori probability of success. The controller does not know or need to know the details of the error recovery procedures.

Secondary storage facility empolying serial communications between drive

In a system including a plurality of mass storage devices at least one of which includes first and second ports, a plurality of controllers and cables coupling the ports to various ones of the controllers and in which each device can only be on-line through one port at a time, state information is sent from the on-line port of each device to a first controller to which it is coupled until a predetermined command from the controller to the on-line port is sent by the first controller. The device responds to the predetermined command by discontinuing the sending of state information from the on-line port to the first controller, and sending a state available signal from the other port to the second controller while maintaining the actual state of said device unchanged, permitting the other controller to interrogate the device as an aid in determining system topology.

Secondary storage facility employing serial communications between drive

In a data processing system including a host computer and a secondary storage system that includes a controller and a mass storage device, the device stores information regarding the physical and logical characteristics of a disk drive associated with the device. In response to a command from the controller, the device produces a signal or signals which provide information to the controller regarding the physical and logical characteristics of the disk drive. The physical and logical characteristics include the drive bit transfer rate and subunit characteristics.

A radial bus for use in a secondary storage subsystem, between a mass storage drive and controller. The bus has four unidirectional bit-serial channels, two for carrying signals from drive to controllers. One channel carries real-time drive state information to the controller; another carries real-time controller state information to the drive. The state information is a sequence of multiplexed bits sent in continuous repetition. Most status variables are represented as a single bit in a specific place in the sequence; the set of status variables defining drive state or controller state, as the case may be, is thus provided by a sequence of bits. When such a bit changes state, a potential change of status has occurred; the change is required to persist some number of repetition of the sequence before the state change is recognized, to avoid spuriously signalling a state change.

A control arrangement for coordinating operations of multiple processors in a multiprocessor system in response to a command. Each command is associated with a route comprising a sequence of route vectors, each identifying an operation to be performed to execute the command, as well as the process, or station, to execute the route vector. In response to the receipt of a command, a control block is generated identifying the first route vector in the route associated with the command. Each station has a work queue containing control blocks, which the station retrieves and processes sequentially. The control block is first sent to the work queue of the station to perform the first operation. When the station gets to the control block, it performs the operation required by the route vector, modifies the control block to identify the next route vector in the sequence, and transfers the control block to the work queue of the station to perform the operation required by the next route vector in the route.

Apparatus and method for controlling digital data processing system

An interface mechanism (10) between two processors, such as a host processor (70) and a processor (31) in an intelligent controller (30) for mass storage devices (40), and utilizing a set of data structures employing a dedicated communications region (80A) in host memory (80). Interprocessor commands and responses are communicated as packets over an I/O bus (60) of the host (70), to and from the communication region (80A), through a pair of ring-type queues (80D) and (80E). The entry of each ring location (e.g., 132, 134, 136, 138) points to another location in the communications region where a command or response is placed. The filling and emptying of ring entries (132-138) is controlled through the use of an `ownership` byte or bit (278) associated with each entry. The ownership bit (278) is placed in a first state when the message source (70 or 31) has filled the entry and in a second state when the entry has been emptied. Each processor keeps track of the rings' status, to prevent the sending of more messages than the rings can hold. These rings permit each processor to operate at its own speed, without creating race conditions and obviate the need for hardware interlock capability on the I/O bus (60).

Growing community of inventors

Colorado Springs, CO, United States of America

Edward A Gardner