Abstract: A priority queue structure and algorithm for managing the structure which in most cases performs in constant time. In other words, most of the time the inventive algorithm performs its work in an amount of time that is independent on the number of priority classes or elements that exist in the queue. The queue itself consists of a linked list of elements ordered into subqueues corresponding to priority classes, with higher priority subqueues appearing earlier in the queue. An array of priority pointers contains an entry for each subqueue that points to the last element of each subqueue. Elements are removed for processing from the top of the queue. Removal takes constant time. Items are inserted into an appropriate subqueue by linking it at the end of its respective subqueue.

Abstract: Controller triple modular redundancy is substantially achieved and reliability improved in a system having duplicate controllers that serve peripheral units. Both controllers detect suspected faults in itself and in the other controller. A peripheral unit that suspects a faulty active controller requests a switch of the active controller. A voting circuit processes votes from the controllers and the active controller switch signal from the peripheral units to select the active controller. The signaling paths between the controllers used to convey votes and active controller information are duplicated. The signals on these signaling paths convey information by using oscillating signals of different frequencies.

Abstract: Management of datastream construction prior to transmission of the datastream across a channel of a communications system by providing for data blocking while reducing movement or copying of the data improves the performance in a communications system. Multiple header segments received from a higher layer in the communications stack are copied into the datastream header area of a datastream such that the header segments are sequentially stored in the datastream header area. A datastream buffer list having entries referencing the datastream header area is generated. Buffer list entries referencing data segments received from higher layers in the communications stack are also stored in the datastream buffer list. The data segments are not physically moved or copied into the datastream during processing by the communications stack. Rather, a "virtual" datastream is generated by the communications stack for transmission without physically moving or copying the data segment.

Abstract: A method of providing a voice mail service is disclosed using a voice processing system having a communications link to one or more voice mail systems. A voice message for a subscriber to said one or more voice mail systems is received at said voice processing system, which prompts a determination as to the voice mail system on which said subscriber has a voice mailbox. The voice message is the transmitted to the voice mail system on which said subscriber has a voice mailbox. In the preferred embodiment, the communications link forms part of the Internet, and received messages are processed independently by two separate voice processing systems to provide redundancy.

Abstract: Multiple virtual routing networks having the same identifier are defined on a shared access transport facility network. Virtual connections are defined from some or all of the network nodes to at least one of the virtual routing networks. A virtual connection contains a medium address for the node to which it is connected. When a route is calculated between a source node and a destination node, the identical identifier representing different virtual routing networks may appear in the calculated route. If it does, all portions of the route after the first occurrence up to and including the last occurrence of the identifier in the calculated route is deleted to achieve the optimal route. A destination node address is obtained from storage that is associated with the virtual connection to the destination node and information can then be directly routed to the destination node address via the calculated route in an efficient manner.

Abstract: Management of the processing of relatively large data objects in a communications stack having multiple layers improves the performance in a communications system in preparing relatively large data objects for transmission across a communications network. This reduces or eliminates data movement and copying during segmentation of the relatively large data objects into relatively small data objects, and appendage of headers to the relatively small data object segments during processing in the communications stack. A shared storage manager creates and controls multiple tokens representing multiple images of portions of the relatively large data objects to enable separate scheduling of the multiple images from the same storage unit or buffer to be passed from one layer in a communications stack to the next lower layer in the communications stack. The large data object is segmented into a plurality of relatively small data object segments at one or more of the layers in the communications stack.

Abstract: Transmission control improves the performance in a communications system relating to transferring large data objects between domains or applications. This reduces or eliminates data movement between domains by transferring ownership of the "container" containing the data as opposed to moving the contents of the container from one domain to another domain resulting in copying the large data object. Thus, transmission control provides for control of the transmission of relatively large data objects between domains in a communications system which otherwise only allows efficient transmission of relatively small data objects between domains in a communications system. Tokens are assigned and associated with buffers containing the data to be shared. The token is not an address, but rather an identifier for the buffer which can be transferred from one domain to a second domain without requiring the copying of the data.

Abstract: A priority queue structure and algorithm for managing the structure which in most cases performs in constant time. In other words, most of the time the inventive algorithm performs its work in an amount of time that is independent on the number of priority classes or elements that exist in the queue. The queue itself consists of a linked list of elements ordered into subqueues corresponding to priority classes, with higher priority subqueues appearing earlier in the queue. An array of priority pointers contains an entry for each subqueue that points to the last element of each subqueue. Elements are removed for processing from the top of the queue. Removal takes constant time. Items are inserted into an appropriate subqueue by linking it at the end of its respective subqueue.

Abstract: A method, apparatus and computer program product for reducing the data transmitted over an external communication link using the TCP protocol from a first application resident in a first computer and to a second application resident in a second computer. The method, apparatus and computer program product include establishing a first virtual socket in the first computer in response to each connection request by the first application for receiving request data originated by the first application. A first real socket in the first computer and a second real socket in the second computer are established to connect the first computer to the second computer over the external communication link and are maintained until the request data has been provided to the second application.

Abstract: A server computer connectable to a network and having a plurality of processors (20, 30, 40) arranged to provide a service to one or more client computers connected to the network. The service Involves the provision of blocks of information to a client computer (50), and the processors have access to a data storage (120) in which data for the blocks of information is stored. Each processor in the server has block retrieval means (80), employed in response to a request from a client computer (50) for a particular block of information from that processor, to retrieve data from the data storage (120) to construct a file representing the requested block of information, the file including references to subsequent blocks of information requestable by the client (50). Additionally, load determining means (100) periodically determines activity data for the processor for inclusion In a load distribution record maintained for all of the processors of the server.

Abstract: A method, apparatus and computer program product for reducing the data transmitted over an external communication link from a first application resident in a first computer to a second application resident in a second computer. The method, apparatus and computer program product include storing a data stream from the first application to be provided to the second application in response to a request from the second application in a cache resident in the first computer to create a server base cache entry and in a cache resident in the second computer to create a client base cache entry. Requests from the second application are interrogated to determine if a client base cache entry corresponding to the interrogated request exists and to determine if a server base cache entry corresponding to the interrogated request exists.

Abstract: Effective memory management maximizes the use of main memory in a computing system by avoiding the issuance of operating system primitives which result in overhead and increased processing time. This allows an applications program to change the state of a storage unit such as a buffer or page without requiring the issuance of an operating system primitive to change the physical state of the storage unit. A storage manager is provided for controlling the movement of data between storage units in secondary storage and storage units in main memory. A storage unit state indicator or flag is associated with each storage unit in main memory. In addition, a system state indicator for indicating the physical state of a storage unit is also associated with each storage unit. The system state indicator may be set to one of the states of fixed or pageable while the storage unit state indicator may be set to one of the states of fixed, pageable or don't care.

Abstract: A multi-path channel interface for computer input-output systems includes the ability to define and activate unbalanced groups of unidirectional communications sub-channels for a user application. Protocol independent exchange identifications permit not only unbalanced transmission groups but also allow user-controlled extensions for negotiating the values of transmission parameters at the time the transmission group is activated. When error correcting re-transmissions force the re-segmenting of data blocks, second level sub-segment indexing assure the proper order of delivery of the various segments and sub-segments. The exchange identifications include an identification of the user protocol being supported and thus permit interfacing with any user protocol.

Abstract: A multi-path channel interface for computer input-output systems includes the ability to define and activate unbalanced groups of unidirectional communications sub-channels for a user application. Protocol independent exchange identifications permit not only unbalanced transmission groups but also allow user-controlled extensions for negotiating the values of transmission parameters at the time the transmission group is activated. When error correcting re-transmissions force the re-segmenting of data blocks, second level sub-segment indexing assure the proper order of delivery of the various segments and sub-segments. The exchange identifications include an identification of the user protocol being supported and thus permit interfacing with any user protocol.

Abstract: An improvement in the Open Systems Interconnection model in which Service Access Points can be established directly between peer entities in the same layer of a protocol stack. This direct seamless peer-to-peer communication is vastly superior to the use of a higher level relay function to set up and manage the communication path. In particular, setup messages are exchanged between the peer entities and a higher layer to create the Service Access Point between the peer entities. The setup and management capabilities are contained in each of the peer entities and hence no relay function is required after the peer Service Access Point, substantially speeding up the communication and reducing the management overhead.

Abstract: The invention allows the participation of a resource manager in a transaction which is coordinated by a transaction manager with which the resource manager's protocols are not compatible. More particularly, a mapper records a connection reference which the application program obtains from the resource manager and a transaction reference which the application program obtains from the transaction manager. The resource manager stores recovery information in an associated log file.

Abstract: A method, apparatus and program product for increasing the performance of a client/server system having a client application resident on a first computer and communicating with a server application resident on a second computer remote from the first computer. The client application and the server application utilize a client/server independent communication protocol for communication between the client and the server and at least one segment of the communication between the client application in the first computer and the server application in the second computer occurs over an external communication link. The method, apparatus and program product intercept communications in the client/server independent communication protocol originated by the remote client prior to transmission of the communications on the external communication link and convert the communication originated by the client to a second client/server specific communication protocol.

Abstract: A method and apparatus for interconnecting first and second networks that use the same protocols for communications by a third network. The third network is defined in the first and second networks as a single virtual communications link of the native type used by the first and second networks. In response to requests from initiating nodes in the first and second networks to communicate with destination nodes in the other of the first and second networks, routing computation services in the first and second networks compute routes through the first and second networks that will connect the originating and destination nodes and that include the virtual link as such a route. The virtual link is also used to remotely support the communication needs of dependent nodes in the first and second networks that do not have the ability to establish communications sessions within the networks on their own.

Abstract: A multi-path channel interface for computer input-output systems includes the ability to define and activate unbalanced groups of unidirectional communications sub-channels for a user application. Protocol independent exchange identifications permit not only unbalanced transmission groups but also allow user-controlled extensions for negotiating the values of transmission parameters at the time the transmission group is activated. When error correcting re-transmissions force the re-segmenting of data blocks, second level sub-segment indexing assure the proper order of delivery of the various segments and sub-segments. The exchange identifications include an identification of the user protocol being supported and thus permit interfacing with any user protocol.

Abstract: A packet communications network includes both packet switched and circuit switched transmission links. A path test command launched from a source resource on the network to a destination resource follows a current path from the source to the destination. At each resource, the path test command initiates a process which identifies the next resource in the path, the port of the current resource through which the next resource is accessed, and the current time. This data is returned to the source resource where the data is accumulated for the entire path and then analyzed to determine the location of failed links, if any, the response time from the source to the destination, the response time of each resource in the path, and the location of congested links. Packet switched resources identify the path test command themselves and carry out the testing procedure. Circuit switched resources utilize the control point controlling that resource to carry out the testing procedure.