Electrical computers and digital processing systems: multicomput – Computer-to-computer data routing – Least weight routing
Reexamination Certificate
1997-08-29
2001-10-30
Banankhah, Majid (Department: 2151)
Electrical computers and digital processing systems: multicomput
Computer-to-computer data routing
Least weight routing
C702S010000
Reexamination Certificate
active
06311226
ABSTRACT:
BACKGROUND OF THE INVENTION
Data communication in a computer network involves the exchange of data between two or more entities interconnected by communication links. The entities are typically software programs executing on hardware computer platforms, such as nodes; in particular, communication software executing on the nodes correlate and manage data communication with other nodes. The nodes typically communicate by exchanging discrete packets or frames of data according to predefined protocols. In this context, a protocol consists of a set of rules defining how the nodes interact with each other. Collectively, these hardware and software components comprise a communications network and their interconnections are defined by an underlying architecture.
Modem communications network architectures are typically organized as a series of hardware and software levels or “layers” within each node that interact to format data for transfer over the network. Predetermined services are performed on the data as it passes through each layer and the layers communicate with each other by means of the predefined protocols. The lower layers of these architectures are typically implemented in hardware and firmware, whereas the higher layers are generally implemented in the form of software running on the nodes attached to the network. Examples of such communications architectures include the the Internet communications architecture and the Systems Network Architecture (SNA) developed by International Business Machines (IBM) Corporation.
SNA is a mainframe-oriented network architecture that includes services generally similar to those defined in the Internet communications architecture. An SNA network consists of nodes and links, wherein the nodes are network components containing protocol implementations and the links are transmission facilities that carry data between two nodes configured to operate a data link control procedure. Examples of such nodes include a host mainframe computer, a control unit and an input/output (I/O) device that provides a user interface to the network. In one embodiment of the SNA network, the control unit and I/O device may be combined within a node, such as a workstation and in another embodiment, the control unit may be independent of the workstation and include a router to enable routing of data through the network to destination nodes, such as workstations.
The host is typically connected to the control unit through a high-performance communication subsystem called a mainframe channel. The channel comprises a plurality of components including an intelligent processor (i.e., channel CPU) that manages the protocol over the communications link and controls transfer of data between host (main memory) storage and I/O devices directly attached to the control unit. To that end, a channel may use one or more channel paths as the actual links between a host and the control unit. Channel paths include physical transmission links between the channel and control unit; examples of channel paths include bus-and-tag and enterprise system connection (ESCON) channel paths. Moreover, each I/O device is represented by a subchannel. A subchannel is similar to a virtual circuit in that it provides information about the associated I/O device and its attachment to the channel.
To transfer data in connection with an I/O operation, the channel CPU executes channel command words (CCWs) once started by a start subchannel operation. The start subchannel command is issued by the host CPU to instruct the channel CPU as to the location of a channel program; this command also specifies the subchannel on which the channel program should execute. The channel program consists of a collection of CCWs; the CCWs are the actual I/O commands (read, write, status, etc) that cause information to flow between the host and an I/O device. The control unit interprets these CCWs and adapts them to fit the characteristics of different I/O devices. Upon issuing a start subchannel operation, the host CPU is released to pursue other processing while the channel organizes the data referenced by the channel program and synchronizes its transfer between the I/O device and main memory.
Communication between a channel and control unit is typically governed by various protocols; a protocol originally developed by IBM Corporation for improving the efficiency of data communication between a host computer and a control unit is the common link access to workstation (CLAW) protocol. In a CLAW environment, the control unit is coupled to a CLAW device, which is typically a software entity executing on a node, such as a workstation. Data communication takes place over a channel via the exchange of data packets between the workstation and host. The CLAW protocol achieves data communication efficiency, in part, by avoiding host CPU interrupts during I/O operations through the continuous execution of channel programs over two subchannels: one for write operations and one for read operations.
Logical links are defined in CCWs for read and write operations to establish multiple logical connections within each subchannel directed to different applications executing on the host and workstation. In fact, a main goal of the CLAW protocol is to enable efficient switching among applications specified by the logical links to facilitate data transfers to appropriate outbound interfaces (e.g., FDDI or Ethernet). Accordingly, the logical links are a way to multiplex within a subchannel.
The CLAW protocol generally defines (i) command codes associated with CCWs and (ii) the order in which those command codes are specified in a CCW chain. A logical link number (0-31) is embedded in the CCW command code, wherein number 0 is reserved for a control link and numbers 1-31 specify application-to-application (data) links. The control link path is part of a read/write subchannel pair dedicated to CLAW protocol communication. The host CPU builds a channel program comprising a chain of CCW data structures in main memory that contain instructions (e.g., read, read header, write, transfer-in-channel (TIC)). These instructions are then executed in accordance with the CLAW protocol.
The CLAW protocol also defines two primary sets of control flows over the control link: system validate/system validate response and connection request sequence. The system validate/system validate response control flows are manifested as message packets that propagate over the control link, passing information as data within a predetermined packet format. Before sending data, a system validate/response message flow occurs over the control link to verify the names of the workstation and host.
The workstation and host names are configuration parameters used to ensure that the host is communicating with a proper workstation. If the workstation (or host) name contained in the system validate message is incorrect, a control application provides a non-match return code in the system validate response message along with the expected name. Upon completion of the system validate/response exchange, the control link (0) has been brought up and the host is “aware” of the workstation to which it is connected. However, logical data links still need to be established for application-to-application data communication.
In order to establish a data link between two communicating applications, a connection request sequence is executed between the host and workstation. A control application resident on the host is typically responsible for initiating the connection request sequence to establish logical links (1-31) for data transfers between applications. A channel adapter within the control unit/workstation controls logical link assignments for applications executing on the workstation. The connection request sequence operates to establish the actual links used for host application-to-workstation application data transfers; once established, various types of data flow over these application-application links (e.g., a Transmission Control Protocol/Internet Protocol (TCP/IP) application has IP
Drummond Mark C.
Hiller Dean
O'Gorman Robert
Rosen Jonathan
Banankhah Majid
Caldwell P
Cesari and McKenna LLP
Cisco Technology Inc.
LandOfFree
Method and apparatus for dynamic link name negotiation does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for dynamic link name negotiation, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for dynamic link name negotiation will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2585614